shiri krebs stanford university abstract
TRANSCRIPT
THE LEGALIZATION OF TRUTH IN INTERNATIONAL FACT-FINDING
Shiri Krebs
Stanford University
ABSTRACT
Do legal judgments influence people’s attitudes and beliefs
concerning contested events? This article builds on studies from three
disciplines – law, psychology, and political science – and employs
experimental methods to shed light on the impact of legal institutions on their
intended audiences. The article identifies a rising ‘legalization of truth’
phenomenon – the adoption of legal discourse to construct and interpret facts
outside the courthouse. It argues that legal truth, while providing a
framework of legal terminology and conventions to analyze and understand
facts, comes with a price tag: it triggers cognitive and emotional biases that
frustrate efforts to disseminate controversial information and to resolve
factual disputes; and it lacks the emotional appeal, participatory value, and
social cues that moral expressions or other types of social truth-telling entail.
To demonstrate the legalization of truth process and to measure its
impact on attitudes and beliefs, this article focuses on the practice of
international fact-finding. In recent years, international fact-finding has
become a dominant response to armed conflicts and political violence around
the world. Lacking compulsory jurisdiction, international fact-finding bodies
have adopted legal discourse, assuming that legal reports uniformly inform
the relevant publics with an authoritative account of what happened and
motivate domestic sanctioning of in-group offenders.
This article challenges both assumptions. Based on two survey-
experiments fielded in 2013 and 2014 on representative samples of 1,000 and
2,000 U.S. nationals, respectively, as well as on an original dataset of U.N.
fact-finding missions, the study demonstrates that three elements of legal
discourse - binary legal judgment, ‘hot’ legal terminology, and legal frame -
Law and International Security Fellow, Stanford Center on International Security and Cooperation [CISAC], J.S.D.
Candidate, Stanford Law School. The author wishes to thank Jenny Martinez, Robert MacCoun, Mariano-Florentino
Cuellar, Mike Tomz, Lee Ross, Dan Ho, Beth Van-Schaack, Gabriella Blum, Beth Simmons, Paul Sniderman, Allen
Winer, Charles Perrow, Karl Eikenberry, Bernadette Meyler, David Sloss, and Alison Renteln for their thoughtful
comments, suggestions and advice. I am also grateful for the comments I received from the participants of the 2016
American Society of International Law [ASIL] Annual Meeting ‘New Voices’ panel, in Washington, DC; the 2016
Harvard Experimental Political Science Conference, in Cambridge, Massachusetts; and the 2014 Northern
California International Law Scholars Workshop at U.C. Hastings, in San Francisco, California. This research
project was made possible thanks to the generous financial support of the Christiana Shi Stanford Interdisciplinary
Award in International Studies [SIGF], the Stanford Laboratory for the Study of American Values [SLAV] and the
CISAC Zuckerman research grant.
Krebs The Legalization of Truth
2
harm the perceived credibility and persuasive value of fact-finding reports:
the legal judgment of the fact-finding report is likely to trigger cognitive
biases and belief polarization; ‘hot’ legal terminology is likely to trigger
emotional biases and reduce the perceived fairness of the report; and the
legal frame appears to be less effective than moral frame in influencing
attitudes on accountability.
The article concludes that international and domestic organizations
should rethink their current design and practice of fact-finding bodies,
acknowledge the limitations of adopting a legal discourse to interpret facts,
and recognize the questionable efficacy of legal fact-finding in influencing
attitudes and beliefs. Accordingly, the goals, structures, and processes of
fact-finding should be reorganized, matching goals with appropriate
structures and processes. By revealing the impact of the legalization of truth
on people’s beliefs and attitudes, this article creates a new framework to
understand the failures and successes of legal fact-finding in particular and
the practice and output of legal institutions more broadly.
Krebs The Legalization of Truth
3
CONTENTS
A. INTRODUCTION ............................................................................................................................ 5
B. THE LEGALIZATION OF TRUTH: THEORY AND HYPOTHESES ...................................................... 12
1. Legal Fact-Finding as an International Strategy ............................................................ 13
2. Challenging the Credibility Assumption: More Information, More Bias ...................... 19
(a) Cognitive biases and receptiveness to fact-finding reports ........................................ 21
(b) Emotional biases and receptiveness to fact-finding reports ....................................... 25
(c) Institutional legitimacy and receptiveness to fact-finding reports ............................. 28
3. Challenging the Mobilization Assumption: Legal Truth Does Not Affect Attitudes .... 29
(a) Naming, shaming and legal framing .......................................................................... 29
(b) Are moral judgments more powerful than legal judgments? ..................................... 31
4. The Legalization of Truth: Problems and Solutions ...................................................... 32
C. EXPERIMENT I: THE EFFECTS OF LEGAL TRUTH ON THE PERCEIVED CREDIBILITY OF FACTS .... 34
1. Design............................................................................................................................. 34
2. Measures......................................................................................................................... 37
3. Results ............................................................................................................................ 38
(a) Legal discourse fails to resolve factual controversies ................................................ 38
(b) The invocation of “war crimes” is counterproductive ............................................... 39
(c) Legal judgment trumps an institution’s favorability .................................................. 40
D. EXPERIMENT II: THE EFFECT OF LEGAL AND MORAL FRAMINGS ON WILLINGNESS TO
SANCTION IN-GROUP OFFENDERS ................................................................................................. 42
1. Design............................................................................................................................. 42
2. Measures......................................................................................................................... 43
3. Results ............................................................................................................................ 43
(a) The legalization of truth and attitude mobilization .................................................... 43
(b) Are there more effective alternatives to legal truth? .................................................. 46
Krebs The Legalization of Truth
4
E. DISCUSSION AND THEORETICAL IMPLICATIONS ...................................................................... 48
1. More Information, More Bias ........................................................................................ 48
2. War Crimes Terminology Decreases Perceived Credibility of Fact-Finding Reports ... 49
3. Legal Judgment Trumps the Issuing Institution’s Favorability ..................................... 50
4. Exploring Alternatives to Legal Discourse .................................................................... 51
F. RECOMMENDATIONS: REDESIGNING INTERNATIONAL FACT-FINDING .................................... 52
1. International Fact-Finding Goals and Purposes ............................................................. 53
2. Alternatives to Legal Discourse ..................................................................................... 56
3. The Kunduz Hospital Bombing: Finding the Facts, Preventing Future Attacks ............ 59
G. CAVEATS AND FUTURE RESEARCH ......................................................................................... 61
H. CONCLUSION .......................................................................................................................... 63
I. APPENDICES ........................................................................................................................... 65
1. U.N. Fact-Finding Missions – Goal Dataset .................................................................. 65
2. Experiments’ Sampling and Demographics ................................................................... 67
3. Vignettes and Questionnaires ......................................................................................... 70
4. Summary of Treatment Effects ...................................................................................... 83
Krebs The Legalization of Truth
5
Write it down. Write it. With ordinary ink on ordinary paper; they weren’t given food, they all died of hunger. All. How many? It’s a large meadow. How much grass per head? Write it down: I don’t know.1
***
What happens to these reports when they are ‘released’? The resources that organizations devote to compiling all this information are not matched by attention to how reports are disseminated or what impact they might have on target audiences.2
A. INTRODUCTION
October 3, 2015, 2:08 a.m.: a U.S. Special Operations AC-130 gunship attacked a Doctors
Without Borders [Médecins Sans Frontières, MSF] hospital in Kunduz, Afghanistan, with heavy
fire. Forty-two people were killed, mostly patients and hospital staff members. Dozens of others
were injured. The main hospital building – the only free trauma care hospital in northern
Afghanistan – was severely damaged and subsequently closed. In the aftermath of the attack on
the hospital, many international organizations, including MSF and various bodies of the United
Nations, called for an international fact-finding investigation to establish the truth and to bring
those responsible to justice.3
In the days and months following the attack several investigations were carried out by the
U.S. military, NATO, the United Nations Assistance Mission in Afghanistan [UNAMA] and
1 WISLAWA SZYMBORSKA, Starvation Camp Near Jaslo, in MIRACLE FAIR: SELECTED POEMS OF WISLAWA
SZYMBORSKA 38 (Joanna Trzeciak trans., 2001). 2 Stanley Cohen, Government Responses to Human Rights Reports: Claims, Denials, and Counterclaims, 18 HUM.
RTS. Q. 517, 518 (1996). 3 For example, U.N. Secretary-General Ban Ki Moon strongly condemned the airstrike and called for a “thorough
and impartial investigation into the attack in order to ensure accountability” (Statement attributable to the
Spokesman for the Secretary-General on attack in Kunduz, UNITED NATIONS (Oct. 3, 2015),
http://www.un.org/sg/statements/index.asp?nid=9095). The U.N. High Commissioner for Human Rights demanded
an investigation and suggested that the attack may amount to a war crime (Kunduz hospital airstrikes “inexcusable”
– Zeid, U.N. HUMAN RIGHTS OFFICE OF THE HIGH COMMISSIONER (Oct. 3, 2015),
http://www.ohchr.org/EN/NewsEvents/Pages/DisplayNews.aspx?NewsID=16564&LangID=E. Human Rights
Watch called the U.S. to establish “an independent panel outside the military chain of command with the aim of
establishing the facts and assessing possible culpability for the strike” (Afghanistan: US Inquiry Must Go Past
Admitting Mistakes, HUMAN RIGHTS WATCH (Oct. 6, 2015), https://www.hrw.org/news/2015/10/06/afghanistan-us-
inquiry-must-go-past-admitting-mistakes. And Doctors Without Borders demanded a “transparent and independent
investigation” (Alissa J. Rubin, Doctors Without Borders Says It Is Leaving Kunduz After Strike on Hospital, N.Y.
TIMES (Oct. 4, 2015), http://www.nytimes.com/2015/10/05/world/asia/doctors-without-borders-says-it-is-leaving-
kunduz-after-strike-on-hospital.html.
Krebs The Legalization of Truth
6
MSF.4 Seven months and four fact-finding reports later, one thing became certain: uncertainty
surrounded almost every aspect of the attack and its consequences. Due to MSF’s high profile, its
international network, and the apparent unlawfulness of striking a functioning hospital,
investigating the attack on the MSF hospital in Kunduz had a strong potential to produce
transparent, consistent, and credible findings. Yet instead of settling the dispute over what
happened, the four reports only refined the controversies and exposed the inherent disbelief and
mistrust between different organizations and communities. The factual findings reported by
UNAMA were challenged by counter-facts produced in the U.S. Central Command Report, and at
the end each side remained committed to its own version of the truth.5 Even basic facts, such as
the time frame of the attack, were just as contested as the sophisticated legal analysis.6 This served
4 U. S. Central Command [Centcom], Summary of the Airstrike on the MSF Trauma Center in Kunduz, Afghanistan
on October 3, 2015; Investigation and Follow-on Actions (Apr. 2016),
https://www6.centcom.mil/FOIA_RR_Files/5%20USC%20552(a)(2)(D)Records/1.%20Airstrike%20on%20the%20
MSF%20Trauma%20Center%20in%20Kunduz%20Afghanistan%20-
%203%20Oct%202015/00.%20CENTCOM%20Summary%20Memo.pdf [hereinafter Centcom Report]; NATO,
Executive Summary: Combined Civilian Casualty [CIVCAS] Assessment of an Airstrike on a Medical Facility in
Kunduz City on 03 October 2015 (Nov. 2015),
https://www.shape.nato.int/resources/3/images/2015/saceur/exec_sum.pdf [hereinafter NATO Report]; U.N.
Assistance Mission in Afghanistan, Human Rights and Protection of Civilians in Armed Conflict: Special Report on
Kunduz Province (Dec. 2015),
https://unama.unmissions.org/sites/default/files/special_report_on_kunduz_province_12_december_2015.pdf
[hereinafter UNAMA Report]; Doctors Without Borders, Initial MSF Internal Review: Attack on Kunduz Trauma
Centre, Afghanistan (Nov. 2015),
http://www.msf.org/sites/msf.org/files/msf_kunduz_review_041115_for_public_release.pdf [hereinafter MSF
Report]. 5 Both UNAMA and MSF expressed an inherent disbelief in the findings of the military investigations. See
UNAMA Report, supra note 4, at 12; Doctors Without Borders, Initial Reaction from MSF to Public Release of US
Military Investigative Report on the Attack on MSF Trauma Hospital in Kunduz, Afghanistan, (Apr. 2016),
http://www.doctorswithoutborders.org/article/initial-reaction-msf-public-release-us-military-investigative-report-
attack-msf-trauma. The UNAMA report explicitly criticized both Centcom and NATO investigations for their lack
of independence and effectiveness. UNAMA Report, supra note 4, at 12. The U.S. and NATO fact-finding reports
seemed coordinated, using almost identical words to describe their identical conclusions: for example, the NATO
report concluded that the attack on the hospital resulted from a “series of human errors, compounded by failures of
process and procedure, and malfunctions of technical equipment,” NATO Report, supra note 4 at 1; and the
Centcom Report concluded, similarly, that the attack resulted from a “combination of human errors, compounded by
process and equipment failures,” Centcom Report, supra note 4, at 1. Therefore, the UNAMA Report called to
establish another independent, impartial, prompt, transparent, and effective investigation of the attack against the
hospital. UNAMA Report, supra note 4 at 12. 6 UNAMA published its Special Report on Kunduz in December 2015. UNAMA Report, supra note 4. The report
adopted the findings elaborated in the MSF Report, concluding that the attack on the hospital lasted about an hour,
from 2:07 a.m. until 3:00 or 3:15 a.m., continuing for at least 40 minutes after the hospital personnel first contacted
U.S. authorities in Afghanistan, at 2:19 a.m., informing them that the hospital was under fire. Id. at 7-8; MSF
Report, supra note 4, at 7. The UNAMA Report concluded that the attack possibly amounted to a war crime – the
conclusions were not final because the U.S. military refused to cooperate with the Mission and did not provide the
necessary information to make firm determinations of responsibility and guilt – and that those involved should be
Krebs The Legalization of Truth
7
as a fertile ground for law professors and legal scholars and practitioners to further the debates and
to intensify and highlight the disagreements over the legal analysis of the applicable norms.7
To close this gap, both MSF and UNAMA reiterated their demand to establish an
international fact-finding mission to investigate the incident.8 The repeated and uniform calls for
an international investigation – whether by the International Humanitarian Fact Finding
Commission (IHFFC) or some other ad-hoc mechanism – are not surprising. In recent years,
international fact-finding has become a dominant response to many crisis situations around the
world. Since the beginning of the twenty-first century, the U.N. alone has dispatched forty-eight
fact-finding missions,9 tasked with responsibility to establish the legal facts by documenting
violations of human rights and international humanitarian law.10 Lacking compulsory jurisdiction,
criminally investigated and prosecuted. See id at 11-12. A few months later, in April 2016, the U.S. Central
Command published its own report, following a thorough internal investigation. Centcom Report, supra note 4. The
Centcom Report produced counter-facts, finding that the attack lasted only 30 minutes, from 2:08 a.m. until 2:38
a.m., and concluding that the ‘tragic errors’ that lead to the attack on the hospital did not amount to a ‘war crime’.
Id. at 3. Moreover, the Centcom Report, like its NATO twin, left many factual questions undecided: it was unable to
conclusively determine how many people were killed in the attack (emphasizing it was unable to verify the numbers
provided by Doctors Without Borders), or what caused the series of errors that led to the hour-long attack on the
trauma center, other than the ‘fog of war.’ Id. at 2, 3. The Centcom Report embraced the uncertainty encountered
during combat operations as the main factor contributing to the tragic course of events. See id. Based on these
counter-fact, the U.S. military adopted administrative and disciplinary measures against the sixteen individuals who
were identified as responsible for the errors. Id. at 4. 7 See, e.g., Peter Margulies, Centcom Report on the Kunduz Hospital Attack: Accounting for a Tragedy of Errors,
LAWFARE (May 2, 2016), https://www.lawfareblog.com/centcom-report-kunduz-hospital-attack-accounting-tragedy-
errors; Alex Whiting, Recklessness, War Crimes, and the Kunduz Hospital Bombing, JUST SECURITY (May 3, 2016),
https://www.justsecurity.org/30871/recklessness-war-crimes-kunduz-hospital-bombing/; Jonathan Horowitz, Why
the US Should Cooperate With Investigations Into the Hospital Bombing, JUST SECURITY (Dec. 23, 2016),
https://www.justsecurity.org/28472/cooperate-investigations-hospital-bombing/; Jens David Ohlin, Was the Kunduz
Hospital Attack a War Crime?, OPINIO JURIS (May 2, 2016), http://opiniojuris.org/2016/05/01/was-the-kunduz-
hospital-attack-a-war-crime/; Kevin Jon Heller, Thoughts on Jens’s Post About the Kunduz Attack, OPINIO JURIS,,
(May 3, 2016), http://opiniojuris.org/2016/05/03/thoughts-on-jens-post-about-the-kunduz-attack/. 8 MSF specifically demanded an international investigation by the International Humanitarian Fact-Finding
Commission (IHFFC). Doctors Without Borders, Kunduz: Initial Reaction to Public Release of U.S. Military
Investigative Report on the Attack on MSF Trauma Hospital (Apr. 29, 2016), http://www.msf.org/en/article/kunduz-
initial-reaction-public-release-us-military-investigative-report-attack-msf-trauma; see also Eve Bring, The Kunduz
Hospital Attack: The Existence of a Fact-Finding Commission, EJIL TALK! (Oct. 15, 2015),
http://www.ejiltalk.org/the-kunduz-hospital-attack-the-existence-of-a-fact-finding-commission/ (encouraging the
international community to turn to the International Humanitarian Fact-Finding Commission to investigate this –
and other – events). UNAMA required an independent, impartial, prompt, transparent, and effective investigation.
See UNAMA Report, supra note 4, at 60.
9 This data is based on qualitative information available at the United Nations Library,
http://libraryresources.unog.ch/factfinding/europe. Quantitative analysis of the data is my own. 10 G.A. RES. A/RES/46/59, Declaration on Fact-finding by the United Nations in the Field of the Maintenance of
International Peace and Security (Dec. 9, 1991). For an elaborate account of U.N. fact-finding missions throughout
the years, see Section B(i) below. For a brief overview of fact-finding mechanisms established by the international
Krebs The Legalization of Truth
8
the fundamental goal of these missions is to persuade relevant audiences to accept their findings
and adopt their recommendations. The assumptions underlying these efforts are that legal fact-
finding bodies are well-suited to settle disputes over ‘what happened,’ and that by producing
credible facts, their findings and conclusions will mobilize domestic accountability measures.
This article challenges both assumptions. It argues that legal fact-finding, while providing
a framework of legal conventions and social discourse to analyze and understand facts, comes with
a price tag: the binary legal judgment is likely to trigger cognitive biases and belief polarization;
the ‘hot’ legal terminology (such as ‘war crimes’) is likely to trigger emotional biases and reduce
the perceived fairness of the report; and the overall legal frame is less likely than other social
frames (including moral frames) to influence attitudes on accountability and to mobilize public
support to prosecute in-group offenders and to compensate out-group victims. Therefore, legal
discourse may not be an optimal choice for influencing attitudes and beliefs concerning contested
events. This price is particularly detrimental for international fact-finding, because, in contrast to
international tribunals, international fact-finding bodies suffer from an enforcement deficit and are
therefore designed to influence their intended audiences through soft power, dialogue, and
persuasion. While MSF understandably desires to find the truth about what happened the day of
the attack on their hospital – and to punish those responsible – this article suggests that yet another
legal fact-finding report is probably not what is needed to resolve the controversy and mobilize
sanctioning.
Indeed, international fact-finding missions have often been at the center of intense
controversies about alleged atrocities and abuses.11 The Rwandan reaction to the 2010 Democratic
Republic of the Congo Report (“dangerous and irresponsible”),12 and the Israeli response to the
community since 1913, see Rob Grace & Claude Bruderlein, Building Effective Monitoring, Reporting, and
Fact-finding Mechanisms 3-9 (April 12, 2012), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2038854. 11 Philip Alston and Sarah Knuckey, The Transformation of Human-Rights Fact-Finding: Challenges and
Opportunities, in THE TRANSFORMATION OF HUMAN RIGHTS FACT-FINDING 3 (Philip Alston & Sarah Knuckey eds.,
2016). In a recent U.N. publication, the High Commissioner for Human Rights welcomed such controversies as
“publicity,” and argued that they increase the public profile of the investigations and thus highlight the important
role of the investigations in promoting accountability. Office of the High Commissioner of Human Rights
[OHCHR], Commissions of Inquiry and Fact-Finding Missions on International Human Rights and Humanitarian
Law Guidance and Practice, at 7, HR/PUB/14/7 (2015),
http://www.ohchr.org/Documents/Publications/CoI_Guidance_and_Practice.pdf [hereinafter UNHCHR Fact-
Finding Guidance]. 12 Statement by the Government of Rwanda on Leaked Draft UN report on DRC, RWANDA NEWS AGENCY (Aug. 27,
2010), http://rnanews.com/politics/4080-statement-by-the-government-of-rwanda-on-leaked-draft-un-report-on-drc-.
Krebs The Legalization of Truth
9
2009 Gaza Report (“false and distorted”),13 are merely two recent examples of the heated
controversies generated by the release of many of the U.N. fact-finding reports.14 Analyzing public
opinion polls in the former Yugoslavia, Marko Milanovic revealed that factual controversies
continue to thrive even when the factual determinations are made by a competent court, whose
judgements are binding and final.15 While the International Criminal Tribunal for the former
Yugoslavia [ICTY] prides itself in “creating a historical record, combatting denial and preventing
attempts at revisionism,”16 Milanovic demonstrated that denialism is thriving, that each ethnic
group in the former Yugoslavia is still firmly attached to its own version of history, and that the
tribunal failed to persuade the relevant target populations that the findings in its judgments are
true.17 Nonetheless, as Milanovic points out, the Yugoslav surveys do not provide us with a direct
measure of any impact that ICTY has had on people’s attitudes about crimes in the Yugoslav wars,
or what the situation would look like had the ICTY done things differently. He concludes,
therefore, that “the best thing we can do is speculate, if in an informed way.”18
This article takes a different methodological approach – one that obviates the need to
speculate. It complements existing observational studies by leveraging the use of experiments
embedded in a pair of U.S. national surveys. Using survey-experiments, this article provides
systematic evidence of the consequences of international fact-finding reports on people’s attitudes
and beliefs. It directly measures individuals’ reactions to both positive and negative information
(sanctioning or absolving in-group members); different legal terminologies (‘cold’ violations
terminology versus ‘hot’ war crimes terminology); and different frames (legal frame versus moral
frame.) Based on these components of legal fact-finding reports – judgment, terminology, and
frame – the article demonstrates how cognitive and emotional biases influence people’s reactions
13 Barak: Goldstone Report “False, Distorted, and Irresponsible,” HAARETZ (Jan. 28, 2010),
http://www.haaretz.com/news/barak-goldstone-report-false-distorted-and-irresponsible-1.265821. 14 Following the release of the 2009 Goldstone Report, an Israeli public opinion poll found that 93.5% of Jewish
Israelis believed the report was biased against Israel. Efraim Ya’ar & Tamar Herman, PEACE INDEX - SEPTEMBER
2009 (2009), http://www.tau.ac.il/peace/. 15 Marko Milanovic, The Impact of the ICTY on the Former Yugoslavia: An Anticipatory Post-Mortem, AM. J. INT’L
L., (forthcoming 2016) (manuscript at 33, 35) (available at
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2755505). 16 U.N. International Criminal Tribunal for the Former Yugoslavia, Achievements,
http://www.icty.org/en/about/tribunal/achievements. 17 Milanovic, supra note 15, at 33, 35. 18 Id. at 28-30.
Krebs The Legalization of Truth
10
to new information about war crimes committed by their fellow nationals. Querying mass publics
also allows me to examine more directly the impact of ‘doing things differently’ – of various
framings, processes, and institutional designs – on willingness to believe threatening facts.
Additionally, using an experimental design minimizes problems posed by selection effects that
characterize much of the existing research on international law and institutions.19 And lastly,
focusing on the social impact of international fact-finding is especially instructive, because, unlike
the ICTY and similar tribunals, fact-finding missions do not give binding judgments or decisions,
and their main contribution is by disseminating facts and making recommendations through
otherwise unenforceable reports.
The two experiments discussed in this article were fielded in 2013 and 2014 on
representative samples of 1,000 and 2,000 U.S. nationals, respectively. The findings demonstrate
that international fact-finding reports on war crimes committed by U.S. Marines in Afghanistan
are ineffective in both (i) resolving controversies in the U.S. over contested events; and in (ii)
motivating domestic condemnation of U.S. war criminals. Additionally, I find that a moral
discourse is more effective than a legal discourse in influencing attitudes on sanctioning and
mobilizing public support to prosecute U.S. war criminals and to compensate Afghan victims.
Based on these findings, I conclude that international organizations, including the U.N.,
should rethink their current design and practice of international fact-finding, acknowledge the
limitations of adopting a legal discourse to interpret facts, and recognize the questionable efficacy
of legal fact-finding in influencing attitudes and beliefs. Accordingly, the goals, structures, and
processes of international fact-finding should be reorganized, matching goals with appropriate
structures and processes: for example, if the main goal of a fact-finding mission is legal
accountability, a court-like structure, complete with enforcement powers, is advisable; if, however,
conflict resolution is the primary goal, then a narrative or restorative approach to truth, through a
social process of truth and reconciliation, is preferable. Ultimately, international fact-finding, in
its current form, carries mainly institutional and social disadvantages.
19 See Adam S. Chilton & Dustin H. Tingley, Why the Study of International Law Needs Experiments, 52 COLUM. J.
TRANSNAT’L L. 173 (2013). For a similar approach, see Geoffrey P.R. Wallace, International Law and Public
Attitudes Toward Torture: An Experimental Study, 67 INT’L ORG. 105 (2013).
Krebs The Legalization of Truth
11
While these findings have concrete implications for the design and practice of international
fact-finding bodies, they also have broader significance concerning the impact of legal institutions
– international or domestic – on their intended audiences, and on the role of legal discourse in
shaping attitudes and beliefs.20 The ‘legalization of truth’ process I identify in this article
characterizes not only the discourse adopted by international organizations, but also the discourse
adopted by many human rights organizations, social movements, and domestic institutions. From
police violence to sexual harassment to torture, the most burning social controversies of our time
have been analyzed in the social sphere through legal lenses, using legal discourse, including
binary legal judgment, legal terminology and legal frames. However, this article demonstrates that
the popularity of legal discourse in fact-finding efforts is probably unjustified and might be
counter-productive. By revealing the impact of legal discourse on people’s beliefs and attitudes,
this article creates a new framework to understand the failures and successes of legal fact-finding
in particular and the practice and output of legal institutions more broadly.
The article begins, in Section B, with an interdisciplinary theoretical framework on the
legalization of truth in international fact-finding. This section combines legal scholarship on war
crimes investigations, fact-finding, and legal discourse with social-psychology literature on
selective information processing, cognitive and emotional biases, attitude formation, and
persuasion; and political science literature on ideological bias and the efficacy of fact-finding
reports. Additionally, this section presents data from an original dataset of U.N. fact-finding
missions, including their mandates and goals. Based on this rich literature and data, I develop my
argument and present my hypotheses concerning the impact of legalization of truth in international
fact-finding on attitudes and beliefs. Sections C and D present the design of the two experiments
and report their findings. Section E then discusses the contribution of these findings to our
understanding of the impact of legal discourse on attitudes and beliefs; and section F offers several
recommendations for the design of fact-finding bodies, including rethinking their goals, and
adopting alternative structures to promote accountability.
20 In a future paper, When More Information Means Less Knowledge: A Comparative Analysis of the U.S., Canada
and Israel (work-in-progress), I will test the generalizability of these findings to other societies, including Canada
and Israel. This forthcoming paper will report the results of experiments fielded in Israel and Canada in 2016. These
comparative experiments – replicating the U.S. experiments, with the necessary changes – will enable me to
compare the impact of international fact-finding reports on factual beliefs and attitudes in different societies.
Krebs The Legalization of Truth
12
B. THE LEGALIZATION OF TRUTH: THEORY AND HYPOTHESES
The ‘legalization of truth’ process refers to the adoption of legal discourse to construct and
interpret facts. Discourse is the foundation of the process of social construction upon which social
reality depends. It relates to the practices of talking and writing, which produce reality or ‘truth’
through the production, dissemination, and consumption of texts.21 Legal discourse is a socially
institutionalized meaning-making process, dictating meanings, interpretations, and ideological
preferences, through a variety of compositional, stylistic, and semantic mechanisms.22 It influences
perceptions, attitudes, and beliefs through a commitment, among other things, to binary legal
judgments, to the closure of controversies, to one true meaning of texts, and to legal rhetoric,
terminology, and voice.23
This article examines the social outcomes of the prevailing legal discourse in international
fact-finding. To operationalize legal discourse and measure its unique influence on social attitudes
and beliefs, I focus on three distinct dimensions of legal discourse which are present in most fact-
finding reports: binary judgments (incriminating or absolving), legal terminology or semantics
(‘war crimes’ versus ‘violations’), and legal frame (referring specifically to ‘law’ or ‘legal
standards’ rather than, for example, ‘moral standards’). This article argues that under some
circumstances, the adoption of legal discourse to produce facts and create meaning, may elicit not
assent but, instead, distrust, resistance, and distortion.24 In the following sections I lay out my
argument concerning the distinctive – and sometimes negative – social impact of legal discourse.
First, I describe the institutionalization of legal discourse in international fact-finding efforts.
Second, I discuss the theoretical foundations of this argument, focusing on two of the main
weaknesses of legal discourse: its limited ability to resolve social controversies and disseminate
contested information, and its bounded efficacy in influencing social attitudes and mobilizing
people for action.
21 Cynthia Hardy, Researching Organizational Discourse, 31 INT’L STUD. MGMT. & ORG. 25, 26-27 (2001). 22 PETER GOODRICH, LEGAL DISCOURSE: STUDIES IN LINGUISTICS, RHETORIC AND LEGAL ANALYSIS 170, 203-204
(1987). 23 Gerald B. Wetlaufer, Rhetoric and Its Denial in Legal Discourse, 76 VA. L. REV. 1545, 1551-1553 (1990). 24 Id. at 1546 (suggesting that lawyers’ style of argument is not uniformly effective).
Krebs The Legalization of Truth
13
1. Legal Fact-Finding as an International Strategy
From Palestine to Lebanon, to Darfur and to Lybia; from Georgia to Ukraine, to Serbia and
to Sri Lanka, one of the most certain facts about conflicts is uncertainty about facts.25 The
disagreement about facts may include the history and roots of the conflict, as well as details
concerning ongoing hostilities. Many times, disputes over facts become so intense they nourish
and intensify the conflict,26 and thus pave the way to further crimes and abuses.27
For decades, the international community has been struggling to find constructive
methods to stop such abuses and to promote justice and accountability.28 One of the most important
and common strategies adopted by many international organizations (including U.N. bodies, such
as the Security Council, the General Assembly, the Human Rights Council, the Secretary-General,
and the High Commissioner for Human Rights) is international fact-finding.29 Already in 1991 the
United Nation’s General Assembly signed the Declaration on Fact-finding by the United Nations
in the Field of the Maintenance of International Peace and Security, urging United Nations organs
to undertake fact-finding activities and to obtain detailed knowledge about the factual
circumstances of any dispute in order to maintain international peace and security.30
To provide a complete analysis and identify the legalization of truth in U.N. fact-finding
missions, I compiled an original dataset which includes all U.N. fact-finding missions from 1945
to 2015. The coding scheme includes the year the fact-finding mission was created (or its mandate
was revised); the U.N. organ establishing the mission; the countries that were the subject of
investigation; and the declared goals of the mission (including investigating violations of
25 Bothe, for example, highlights the importance of ascertaining facts in the context of all kinds of conflicts, be they
of a social, political, or legal character. Michael Bothe, Fact-finding as a Means of Ensuring Respect for
International Humanitarian Law, in INTERNATIONAL HUMANITARIAN LAW FACING NEW CHALLENGES 249, 249-267
(Wolff Heintschel Heinegg & Volker Epping eds., 2007). 26 For a detailed discussion of the disagreement over facts as a socio-psychological barrier to conflict resolution, see
Suzanne Retzinger & Thomas Scheff, Emotion, Alienation, and Narratives: Resolving Intractable Conflict, 18
MEDIATION Q. 71 (2000); Peter T. Coleman, Characteristics of Protracted, Intractable Conflict: Toward the
Development of a Metaframework-I, 9 PEACE & CONFLICT 1, 19 (2003). 27 For a comprehensive analysis of the preservation of conflicting narratives and beliefs in intractable conflict, see
Daniel Bar-Tal, Sociopsychological Foundations of Intractable Conflicts, 50 AM. BEHAV. SCIENTIST 1430 (2007);
See also Neta Oren & Daniel Bar-Tal, The Detrimental Dynamics of Delegitimization in Intractable Conflicts: The
Israeli-Palestinian Case, 31 INT'L J. INTERCULTURAL REL. 111 (2007). 28 This paper does not deal with the few international mechanisms which enjoy mandatory jurisdiction and
enforcement power, but only with voluntary fact-finding mechanisms that depend on states’ cooperation. 29 UNHCHR Fact-Finding Guidance, supra note 11, at 2; see also supra notes 1-6. 30 G.A. Res. A/RES/46/59, supra note 10.
Krebs The Legalization of Truth
14
international law, promoting accountability, assisting domestic authorities, promoting
reconciliation, preventing future atrocities, determining facts, and ascertaining the truth). The
dataset is based on information which is available online at the U.N. Library in dozens of
previously unorganized and understudied documents, including U.N. resolutions, reports, and
decisions. The following paragraphs introduce the main findings of this analysis.
Throughout its long history, the United Nations, through its various organs, has established
66 fact-finding missions. Of these 66 missions, 38 (57%) were established in the last decade. The
U.N. Human Rights Council [UNHRC], a U.N. organ established in 2006, is responsible for the
establishment of 29 (44%) of these 66 missions; UNHRC missions comprise 76% of the missions
established in the last decade. Overall, about half of all missions were sent to investigate atrocities
in Africa, and the other half concentrated in Asia and the Pacific. Four missions were tasked with
investigating atrocities in the Americas and one was established to investigate atrocities in the
former Yugoslavia.31
Figure 1. U.N. Fact-Finding Missions, by Decade
Figure displays the number of fact-finding missions established by U.N. organs, by decade.
31 This data is based on qualitative information available at the United Nations Library,
http://libraryresources.unog.ch/factfinding/europe. Quantitative analysis of the data is my own.
0 1 2 1
7
17
38
0
5
10
15
20
25
30
35
40
1946-1955 1956-1965 1966-1975 1976-1985 1986-1995 1996-2005 2006-2015
Number of Fact-Finding Missions Established by UN Organs, by Decades
Krebs The Legalization of Truth
15
In recent decades, the variety of fact-finding mechanisms has tremendously expanded, not
only in numbers, but also in mandates, capacities, institutional affiliations, and goals.32
Nonetheless, while the 1991 General Assembly resolution defined fact-finding in terms of
“obtaining knowledge of the relevant facts,”33 most U.N. fact-finding missions were not created
to fulfill the goal of ascertaining facts alone. Instead, these fact-finding missions were specifically
tasked with other goals, such as determining state and individual responsibility for violations of
international law, making recommendations regarding reform and reparations, and promoting
accountability.34 The mandate of the most recent UNHCR fact-finding mission – the Burundi
Mission – does not even mention ascertaining facts as one of its purposes, and instead focuses on
long-term goals such as preventing further deterioration of the human rights situation, making legal
determinations on possible violations of international law, assisting the state to fulfil its human
rights obligations, and ensuring accountability for human rights violations and abuses.35
Analyzing the mandates of these 66 fact-finding missions, I found that an overwhelming
majority of these missions – 63 of the 66 missions – were established to investigate alleged
violations of international law, requesting the mission members to rely on legal norms in their
analysis and to adopt legal discourse in the description of their findings. Half of those missions –
33 missions – were specifically instructed to further the goal of accountability. Of the 38 missions
established after 2006, 24 (63%) were specifically tasked with the goal of promoting
accountability, while of the 28 missions established prior to 2006, only 9 (32%) were instructed to
focus on accountability. Only three missions were instructed to promote the goal of reconciliation;
an additional three missions were instructed to assist domestic authorities; and the goal of only one
mission was to prevent future atrocities. Of the 66 fact-finding missions, only 17 mandates
included reference to “fact-finding” in their mandate, and the mandate of one mission alone
included reference to ascertaining the “truth.”36 It is clear from this data that the past decade has
32 For a brief overview of fact-finding mechanisms established by the international community since 1913,
see Grace & Bruderlein, supra note 10, at 3-9. 33 G.A. Res. A/RES/46/59, supra note 10. 34 Specifically, these goals include assessing the truth of allegations made; preparing reports designed for “name and
shame” advocacy or as a basis for future litigation; constructing a historical record; and making recommendations to
prevent future abuse. Alston & Knuckey, supra note 11, at 9. 35 U.N. HUMAN RIGHTS COUNCIL Res. A/HRC/RES/S-24/1, Preventing the Deterioration of the Human Rights
Situation in Burundi (Dec. 17, 2015), http://www.ohchr.org/EN/HRBodies/HRC/UNIIB/Pages/UNIIB.aspx. 36 See Appendix I.1 for the full dataset of U.N. fact-finding mission goals.
Krebs The Legalization of Truth
16
brought a meaningful rise not only in the number of fact-finding missions, but also in the adoption
of legal discourse and concrete legal goals – such as accountability.37 The data further suggest that
the process of establishing fact-finding missions reproduces ‘legalization of truth’ approach to
fact-finding and a commitment to legal discourse without questioning its efficacy and without
considering alternatives.38
This legalization of truth in U.N. fact-finding missions echoes the ‘naming and shaming’
strategy adopted by many non-governmental organizations (NGOs).39 ‘Naming and shaming’ is a
fact-finding process designed specifically for the goal of exposing concrete violations of
international humanitarian law and international human right law (‘naming’), and publicizing this
information to condemn the relevant societies (‘shaming’), in the hope that the international
condemnation will pressure the relevant societies to stop the abuses and bring the offenders to
justice.40 Fact-finding efforts designed to ‘name and shame’ abusive regimes typically convey their
messages through legalized fact-finding reports, concluding with semi-judicial determinations that
international law was (or was not) violated. There are various theories explaining why ‘naming
37 The qualitative data further demonstrate the nature of this legalization of truth. For example, the language of the
UNHRC resolution mandating a fact-finding mission in Burundi extended the mission’s mandate to investigate
“violations and abuses of human rights,” UNHRC Res. S-24/1, supra note 35, ¶ 17(a). Other recent examples are the
Security Council fact-finding mission to the DRC, which defined the mission’s mandate as “a mapping of the
serious violations committed,” S.C. Res. S-2006/390, ¶ 54 (June 13, 2006); and the mandate of the 2009 Gaza
mission, which called to “investigate all violations of international human rights law and international humanitarian
law,” UNHRC Res. A/HRC/S-9/1, 153-156. See also G.A. Res. A/RES/68/165, ¶ 1, (Dec. 18, 2013) (stating that the
organization “recognizes the importance of respecting and ensuring the right to the truth so as to contribute to
ending impunity and to promote and protect human rights”); UNHRC Res. A/HRC/RES/14/7, Proclamation of 24
March as the International Day for the Right to the Truth Concerning Gross Human Rights Violations and for the
Dignity of Victims (June 23, 2010); Frédéric Mégret, Do Facts Exist, Can They Be “Found”, and Does It Matter?,
in THE TRANSFORMATION OF HUMAN RIGHTS FACT-FINDING, supra note 11, at 34. 38 For an elaborate criticism of the design process of international fact-finding mechanisms, see Q.C. Palmer,
Reform of UN Inquiries, in FOR THE SAKE OF FUTURE GENERATIONS – ESSAYS ON INTERNATIONAL LAW, CRIME,
AND JUSTICE IN HONOUR OF ROGER S. CLARK, 597, 610-615 (2015). See also Grace & Bruderlein, supra note 10
(calling on the international community to adopt a more systematic and informed process for establishing fact-
finding mechanisms). 39 For example, Human Rights Watch declares that its mission is to “investigate abuses, expose the facts widely, and
pressure those with power to respect rights and secure justice,” About, HUMAN RIGHTS WATCH,
https://www.hrw.org/about, and states that the organization’s founding strategy was to adopt “a methodology of
publicly ‘naming and shaming’ abusive governments through media coverage and through direct exchanges with
policymakers,” Our History, HUMAN RIGHTS WATCH, https://www.hrw.org/our-history. See also, Emilie M. Hafner-
Burton, Sticks and Stones: Naming and Shaming the Human Rights Enforcement Problem, 62 INT’L ORG. 689
(2008); Suzanne Katzenstein, Reverse-Rhetorical Entrapment: Naming and Shaming as a Two-Way Street, 46
VAND. J. TRANSNAT'L L. 1079, 1079 (2013). 40 Amanda M. Murdie & David R. Davis, Shaming and Blaming: Using Events Data to Assess the Impact of Human
Rights INGOs, 56 INT’L STUD. Q. 1, 56 (2012); Katzenstein, supra note 39, at 1079.
Krebs The Legalization of Truth
17
and shaming’ works,41 but all rest on two fundamental assumptions: (a) that fact-finding reports
resolve factual controversies concerning ‘what happened’ (which I refer to as the ‘credibility
assumption’);42 and (b) that publicizing these reports pressures the relevant societies to stop the
abuses and sanction the offenders (which I refer to as the ‘mobilization assumption’).43
While the legal literature on the effects of fact-finding is underdeveloped,44 political
science scholars have examined its efficacy through various methods, both quantitative and
qualitative. The data is mixed: while some scholars (and many international organizations) believe
that fact-finding leads to decreased violations and promotes justice and accountability,45 others
argue that this strategy may prove to be counter-productive and prevent political compromises that
could otherwise lead to peace and stability.46 The debate is far from being settled,47 and some
scholars find that the answer is complex, and that the efficacy of this strategy depends upon many
domestic and international factors, including the type of shamed regime,48 the type of abuses,49
and the identity of the condemning body.50 Additionally, existing data is limited in scope and
focuses on the impact of naming and shaming on governments and elites, or on levels of violence
41 Matthew Krain, J’accuse! Does Naming and Shaming Perpetrators Reduce the Severity of Genocides or
Politicides?, 56 INT’L STUD. Q. 574 (2012); James C. Franklin Shame on You: The Impact of Human Rights
Criticism on Political Repression in Latin America, 52 INT’L STUD. Q. 187 (2008). 42 Davis, Murdie, and Steinmetz empirically tested and confirmed the hypothesis that Human Rights Organizations
[HROs] are powerful conduits through which a population becomes informed of domestic human rights issues.
David R. Davis, Amanda Murdie & Coty Garnett Steinmetz, ‘Makers and Shapers’: Human Rights INGOs and
Public Opinion, 34 HUM. RTS. Q. 199 (2012). More generally, Wallace summarizes the literature on the ‘educational
role’ of international law as a tool to communicate messages to individuals. Wallace, supra note 19, at 111. See also
Krain, supra note 41. 43 Jacqueline H. R. DeMeritt, International Organizations and Government Killing: Does Naming and Shaming
Save Lives?, 38 INT’L INTERACTIONS 597 (2012); Krain, supra note 41. 44 In recent years, a scholarly literature describing and analyzing international fact-finding has developed steadily.
Nonetheless, as Alston and Knuckey rightly identify, the vast majority of these studies are either essentially
descriptive or largely uncritical. Alston & Knuckey, supra note 10, at 4. 45 See, e.g., Thomas Risse & Kathryn Sikkink, The Socialization of International Human Rights Norms into
Domestic Practices: Introduction, in THE POWER OF HUMAN RIGHTS: INTERNATIONAL NORMS AND DOMESTIC
CHANGE 1,5 (Thomas Risse, Stephen Ropp, & Kathryn Sikkink eds., 1999); E. M. Hafner-Burton & K. Tsutsui,
Human Rights in a Globalizing World: The Paradox of Empty Promises, 110 AM. J. SOCIOLOGY 1373 (2005). 46 See, e.g., Jack Snyder & Leslie Vinjamuri, Trials and Errors: Principle and Pragmatism in Strategies of
International Justice, 28 INT’L SECURITY 5, 6 (2004); Franklin, supra note 41, at 187 (presenting a more nuanced
argument and claiming that the positive impact of naming and blaming is temporary). 47 Katzenstein, supra note 39. 48 Anja Jetschke & Andrea Liese, The Power of Human Rights a Decade After: From Euphoria to Contestation?, in
THE PERSISTENT POWER OF HUMAN RIGHTS: FROM COMMITMENT TO COMPLIANCE 26 (Thomas Risse, Stephen
Ropp, & Kathryn Sikkink eds., 2013). 49 Hafner-Burton, supra note 39. 50 Franklin, supra note 41.
Krebs The Legalization of Truth
18
and abuses.51 Very little work has been done to test the impact of fact-finding reports on the
investigated societies more broadly or to directly examine either the credibility or mobilization
assumptions.
No doubt it is important to pressure governments and oppressive regimes directly
(especially when dealing with non-democratic regimes). Nonetheless, the international community
should acknowledge the fact that some of the countries that engage in activities that violate
international law are democracies, and that public opinion in those countries is important for their
collaboration with non-binding international reports and institutions. In various studies, political
scientists have demonstrated that public opinion affects policy in a variety of fields, and that such
influence is well-maintained even in the face of activities by interest organizations, political
parties, and political and economic elites.52 Moreover, the focus on governments and elites misses
one of the most important aspects of international fact-finding: the prospect of providing
conflicting societies with credible information on contested events, and creating a shared
understanding of ‘what really happened.’ Can fact-finding be considered successful if it triggers
more conflict and disagreements over what happened? Can it be considered effective if it fails to
persuade the perpetrators’ society that offenders should be condemned and sanctioned?
A recent article published by the Journal of Peace Research took on the task of empirically
testing the impact of ‘naming and shaming’ strategies on individuals. The article, by Jacob
Ausderan, concluded that after being exposed to negative information on abuses conducted by
their governments, individuals updated their beliefs accordingly to reflect more negative
attitudes.53 In other words, as expected by proponents of the ‘naming and shaming’ strategy, more
information means more shared knowledge. Ausderan, however, acknowledged the potential of
positive reaction to negative information, and admitted that further research is necessary to shed
51 BETH SIMMONS, MOBILIZING FOR HUMAN RIGHTS: INTERNATIONAL LAW IN DOMESTIC POLITICS (2009);
JACK L. GOLDSMITH & ERIC A. POSNER, THE LIMITS OF INTERNATIONAL LAW (2006); Cohen, supra note 2;
DeMeritt, supra note 43; Jack Snyder & Leslie Vinjamuri, supra note 46 . 52 Benjamin I. Page & Robert Y. Shapiro, Effects of Public Opinion on Policy, 77 AM. POL. SCI. REV. 175-190
(1983); Paul Burstein, The Impact of Public Opinion on Public Policy: A Review and an Agenda, 56 POL. RES. Q.
29, 29 (2003). For a comprehensive survey of the impact of public opinion on decision of the U.S. Supreme Court,
see Lee Epstein & Andrew D. Martin, Does Public Opinion Influence the Supreme Court? Possibly Yes (But We’re
Not Sure Why), 13 U. PENN. J. CONST. L. 263 (2010). 53 Jacob Ausderan, How Naming and Shaming Affects Human Rights Perceptions in the Shamed Country, 51 J.
PEACE RES. 81 (2014).
Krebs The Legalization of Truth
19
more light on this issue.54 Additionally, this main effect, which was detected by his analysis of
observational data (limited in its ability to constitute a direct test of the causal mechanism – that
perceptions are affected by individual exposure to naming and shaming reports), was not supported
by the results of his experimental data, which were statistically insignificant. Finally, the article
did not test directly either one of the fundamental assumptions underlying the naming and shaming
strategy: the credibility assumption and the mobilization assumption.55
This article takes on this challenge. It tests both the credibility and mobilization
assumptions underlying international fact-finding efforts, and provides data that reveal how people
process and respond to new information concerning war crimes committed by their fellow
nationals. It directly tests individuals’ reactions to both positive and negative information, and
demonstrates how ideological beliefs and political commitments influence people’s reactions to
new information about war crimes. As a first step, this article begins with the U.S. society and uses
the war in Afghanistan as its context. In a separate article, I will present comparative evidence on
the impact of legalization of truth on attitudes and beliefs in the U.S., Israel and Canada.
2. Challenging the Credibility Assumption: More Information, More Bias
Fact-finding is defined as a way to ascertain facts by way of gathering immediate, credible,
and first-hand information.56 The credibility of a fact-finding report is therefore a central part of
its core definition and it encompasses its methodology,57 recourses,58 and personnel.59
54 Id. at 83. 55 In a comprehensive article surveying the literature on credibility and persuasion, Chanthika Pornpitakpan
highlights the need for studies that investigate the effect of negative initial dispositions on perceived credibility of
new information and persuasion. Chanthika Pornpitakpan, The Persuasiveness of Source Credibility: A Critical
Review of Five Decades' Evidence, 34 J. APPLIED SOC. PSYCHOL. 243, 270 (2004). 56 H. VICTOR CONDÉ, A HANDBOOK OF INTERNATIONAL HUMAN RIGHTS TERMINOLOGY 88, 127 (2004); Bertrand G.
Ramcharan, Introduction, in INTERNATIONAL LAW AND FACT-FINDING IN THE FIELD OF HUMAN RIGHTS IX
(Bertrand G. Ramcharan ed., 2014). 57 Diane F. Orentlicher, Bearing Witness: The Art and Science of Human Rights Fact-Finding, 3 HARV. HUM. RTS. J.
83, 85 (1990); Théo Boutruche, Credible Fact-Finding and Allegations of International Humanitarian Law
Violations: Challenges in Theory and Practice, 16(1) J. OF CONFLICT & SECURITY LAW 105. 58 M. Cherif Bassiouni, Appraising UN Justice-Related Fact-Finding Missions, 5 WASH. U. J.L. & POL’Y 35, 42
(2001). 59 Ramcharan, supra note 59, at 42. Indeed, with regard to the experience of fact-finding it is considered that fact-
finding requires one to perfect strategies for information gathering, interpersonal communication, and analysis of
both the legal and non-legal dimensions of a particular socio-legal problem. Johanna Bond, The Global Classroom:
International Human Rights Fact-Finding as Clinical Method, 28 WM. MITCHELL L. REV. 317, 328 (2001).
Krebs The Legalization of Truth
20
International fact-finding rests on the basic assumption that more information means more
credible – and thus shared – knowledge, and that by disseminating their findings fact-finding
missions combat denial and resolve controversies concerning disputed facts.60 Fact-finders are
regarded as ‘the eyes and ears’ of the relevant communities,61 and there is a sense that ‘the report
speaks for itself.’62 However, facts very rarely speak for themselves. In his book, States of Denial,
influential sociologist Stanley Cohen explains how individuals select information that fits their
existing perceptual frames, while information which is inconsistent with their existing beliefs or
which is too threatening is shut out altogether.63 Cohen further develops the concept of interpretive
denial, where the raw facts are accepted, but are given a different meaning from what seems
apparent to others.64 In Cohen’s words, “a perceptual filter is placed over reality” and some
knowledge has to be rejected or be given a different meaning.65
The perceived credibility of fact-finding reports may be affected by several aspects of the
reports: its message (the binary legal judgment);66 its terminology (the legal labels used to describe
the relevant behavior, such as ‘war crimes’);67 and its source (the institution producing the report
or establishing the fact-finding mission, such as the U.N. and its organs, national or transnational
NGOs, or domestic institutions, including militaries).68 As credibility is typically assumed, or
treated as a prerequisite of fact-finding mechanisms, it is rarely discussed or studied as a dependent
variable, influenced by the message content, its language, delivery, and source.69 Since the
fundamental goal of any fact-finding institution is to produce a credible description of specific
events, the thriving legal literature on international fact-finding could benefit from lessons learned
60 The High Commissioner for Human Rights, Zeid Ra’ad Al Hussein, emphasized that “the public nature of these
reports is important for them to contribute to the historical recording of events, strengthen the calls for
accountability and promote implementation of the recommendations.” UNHCHR Fact-Finding Guidance, supra note
11, at 92. Similarly, Similarly, Christof Heyns, U.N. Special Rapporteur on extrajudicial, summary, or arbitrary
executions and member of the U.N. Independent Investigation on Burundi, stated that “it is crucial to ascertain the
disputed facts in an indisputable manner.” Enhanced Interactive Dialogue on Burundi, HUMAN RIGHTS COUNCIL
(Mar. 22, 2016), http://libraryresources.unog.ch/c.php?g=462695&p=3162785#21493049. 61 Grace & Bruderlein, supra note 10, at 31. 62 Rob Grace, Communication and Report Drafting in Monitoring, Reporting, and Fact-finding Mechanisms, 13
(July 2014), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2462590. 63 STANLEY COHEN, STATES OF DENIAL: KNOWING ABOUT ATROCITIES AND SUFFERING 6 (2013). 64 Id. at 7-8. 65 Id. at 13. 66 See section B(2)(a), infra. 67 See section B(2)(b), infra. 68 See section B(2)(c), infra. 69 See, e.g., UNHCHR Fact-Finding Guidance, supra note 11.
Krebs The Legalization of Truth
21
through decades of intense research on perceived credibility of new information in other
disciplines, including social psychology, political science, sociology, and communications.
The following sections describe cognitive and emotional biases influencing the
assimilation and perceived credibility of new information. These sections does not offer a one-
size-fits-all social response mechanism, but rather suggest that fact-finding reports trigger various
biases which influence the way information is being processed and perceived. Additionally,
different cognitive and emotional biases may have opposite effects on individuals and groups. For
example, ‘hot’ war crimes terminology may trigger defensive reactions and rejection of
information blaming one’s nation with the commission of war crimes, while holding a liberal
ideology could motivate individuals to question anything that looks like an attempt to whitewash
atrocities.
(a) Cognitive biases and receptiveness to fact-finding reports
Research in the fields of social psychology and communications examines how message
characteristics, including structure, content, and language, impact perceptions of believability of
the message.70 Scholars of information processing have found that the use of opinionated language
– which is perceived as domineering and intolerant – can decrease the credibility of a message,
depending on the legitimacy of the source and the ability of the audience to cope with potential
threat.71 Message discrepancy (the distance between the message position and the pre-message
position of the receiver) is important to understanding the relationship between opinionated
language and credibility:72 discrepant messages, and those that use more intense and opinionated
language, are rated most negatively in terms of credibility.73
Related to message discrepancy, there is evidence that receivers are likely to believe
messages that reaffirm existing beliefs.74 Social-psychology studies have long ago demonstrated
70 Miriam J. Metzger et al., Credibility for the 21st Century: Integrating Perspectives on Source, Message, and
Media Credibility in the Contemporary Media Environment, 27 COMMUNICATIONS YEARBOOK 293, 302 (2003). 71 Mark A. Hamilton, Message Variables that Mediate and Moderate the Effect of Equivocal Language on Source
Credibility, 17 J. LANGUAGE & SOC. PSYCHOL. 109, 116 (1998). 72 Generally, evaluations of credibility are higher when message discrepancy is low. Metzger, supra note 67, at 303. 73 Id.; Hamilton, supra note 68, at 136-137. 74 This is “the illusory-truth effect.” Metzger, supra note 67, at 303 (citing Ian Maynard Begg et al., Dissociation of
Processes in Belief: Source Recollection, Statement Familiarity, and the Illusion of Truth, 121 J. EXPERIMENTAL
PSYCHOL. 446 (1992); see also Alice Dechêne et al., The Truth About the Truth: A Meta-analytic Review of the
Truth Effect, 20(10) PERSONALITY AND SOC. PSYCHOL. REV. 1 (2009).
Krebs The Legalization of Truth
22
that individuals tend to search and absorb information that is in line with their core social beliefs,
while omitting or distorting contradictory information.75 The construction and evaluation of
information in social settings is influenced by the prior beliefs, ideologies, and interests of those
involved,76 as well as their group identities and commitments.77 In heterogeneous societies, in
which different groups hold conflicting social beliefs and ideologies, new information and its
specific framing might trigger belief perseverance or polarization.78 Perceptions of facts are
pervasively shaped by individuals’ commitments to shared but contested sets of values, beliefs,
and identities. Therefore, reaction to new facts or new information is shaped by values-motivated
cognition and identities.79 In a seminal study from 1979, Lord, Ross, and Lepper demonstrated that
exposure to empirical evidence does not typically narrow the gap between those who hold opposed
views at the outset. In fact, exposure to such evidence was found to make initial disagreements
even more pronounced.80 In his book The Righteous Mind – Why Good People are Divided by
Politics and Religion, Jonathan Haidt explains that:
Once people join a political team, they get ensnared in its moral matrix. They see
confirmation of their grand narrative everywhere, and it’s difficult – perhaps
75 Among the various psychological mechanisms which contribute to biased assimilation of information are:
cognitive consistency and confirmation bias, Lee Ross & Andrew Ward, Psychological Barriers to Dispute
Resolution, in 27 ADVANCES IN EXPERIMENTAL SOCIAL PSYCHOLOGY 255, 263-64 (Mark P. Zanna & James M.
Olson eds., 1995); motivated cognition, Dan Kahan, Foreword: Neutral Principles, Motivated Cognition, and Some
Problems for Constitutional Law, 125 HARV. L. REV. 1, 19 (2012); and threatened social identities, Terrell A.
Northrup, The Dynamics of Identity in Personal and Social Conflict, in INTRACTABLE CONFLICTS AND THEIR
TRANSFORMATION 55-82 (Louis Kriesberg et al. eds., 1989); Anne Maass & Mark Schaller, Intergroup Biases and
the Cognitive Dynamics of Stereotype Formation, 2 EUR. REV. SOC. PSYCHOL. 189 (1991); Bar-Tal, supra note 27,
at 1445-1446. 76 Ifat Maoz et al., Reactive Devaluation of an “Israeli” vs. “Palestinian” Peace Proposal, 46 J. CONFLICT RESOL.
515, 543 (2002); see also Roy F. Baumeister & Stephen Hastings, Distortions of Collective Memory: How Groups
Flatter and Deceive Themselves, in COLLECTIVE MEMORY OF POLITICAL EVENTS: SOCIAL PSYCHOLOGICAL
PERSPECTIVES 277, 287 (James W. Pennebaker et al. eds.,1997). Chan, Burtis, and Bereiter's study on knowledge
construction found that individuals distort and twist information to make it fit with prior beliefs. Carol Chan, Jud
Burtis & Carl Bereiter, Knowledge Building as a Mediator of Conflict in Conceptual Change, 15 COGNITION &
INSTRUCTION 1, 5 (1997). 77 Dan M. Kahan et al., They Saw a Protest: Cognitive Illiberalism and the Speech-conduct Distinction, 64 STAN. L.
REV. 851, 859 (2012). 78 Thomas Kelly, Disagreement, Dogmatism, and Belief Polarization, 105 J. PHIL. 611 (2008). 79 Dan M. Kahan, David A. Hoffman & Donald Braman, Whose Eyes Are You Going to Believe? Scott v. Harris and
the Perils of Cognitive Illiberalism, 122 HARV. L. REV. 837, 842-43, 879 (2009). 80 Charles Lord, Lee Ross & Mark Leper, Biased Assimilation and Attitude Polarization: The Effects of Prior
Theories on Subsequently Considered Evidence, 37 J. PERSONALITY & SOC. PSYCHOL. 2098 (1979).
Krebs The Legalization of Truth
23
impossible – to convince them that they are wrong if you argue with them from
outside of their matrix.81
Political science literature has explained the current ‘cultural war’ in the U.S. relaying on these
message discrepancy and belief polarization or perseverance theories.82 By now, it is well-
established that partisans and ideologues in the U.S. express differences in factual beliefs.83
Ideological partisanship affects both how the public learns new information, and which
information it learns.84 Taber and Lodge demonstrated that strong partisans make every effort to
maintain their existing opinions by seeking out confirming evidence, counter-arguing information
that does not fit their preexisting conceptions, and attributing more strength to arguments that
match their opinions.85 The impact of political identification on biased assimilation of information
has been confirmed by Shapiro and Bloch‐Elkon regarding a number of foreign-policy and national
security issues.86 For example, they have demonstrated a sharp gap between Democrats and
Republicans on various factual beliefs relating to the war in Iraq, including whether Iraq had
possessed WMDs and whether clear evidence had been found that Iraq had been supporting al
Qaeda.87 Another study emphasized that even when partisans hold similar factual beliefs, they
interpret this information differently, and these interpretations drive their opinions.88 More
specifically, the study, conducted by Gaines et al., demonstrated that both Democrats and
Republicans updated their factual beliefs about the war in Iraq as conditions changed, but
interpreted the same factual beliefs quite differently. Democrats consistently interpreted a given
level of troop casualties as higher than Republicans did. Similarly, Democrats interpreted the
81 JONATHAN HAIDT, THE RIGHTEOUS MIND: WHY GOOD PEOPLE ARE DIVIDED BY POLITICS AND RELIGION 312
(2012). 82 Dan M. Kahan et al., The Second National Risk and Culture Study: Making Sense Of - and Making Progress In -
the American Culture War of Fact (G.W.U. Legal Studies Research Paper No. 370, 2007),
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1017189. 83 Brian J. Gaines et al., Same Facts, Different Interpretations: Partisan Motivation and Opinion on Iraq, 69 J. POL.
957 (2007); John G. Bullock et al., Partisan Bias in Factual Beliefs About Politics (Nat’l Bureau of Econ. Research,
Working Paper No. 19080, 2013). 84 Robert Y. Shapiro & Yaeli Bloch‐Elkon, Do the Facts Speak for Themselves? Partisan Disagreement as a
Challenge to Democratic Competence, 20 CRITICAL REV. 115, 116 (2008). 85 Charles S. Taber & Milton Lodge, Motivated Skepticism in the Evaluation of Political Beliefs, 50 AM. J. POL. SCI.
755, 767 (2006). 86 Shapiro & Bloch‐Elkon, supra note 81, at 123-24. 87 Id. 88 Gaines et al., supra note 80.
Krebs The Legalization of Truth
24
failure to find weapons of mass destruction in Iraq as evidence that they never existed, while
Republicans inferred that the Iraqi leadership had moved, destroyed, or hidden the weapons.89
For several decades now, party identification in the U.S. has been strongly correlated with
ideological preferences and beliefs.90 Specifically, people have sorted into the party-ideology
combination of liberal Democrats and conservative Republicans.91 As early as 1998, Abramowitz
and Saunders demonstrated that voters have been choosing their party identification on the basis
of their ideological preferences rather than maintaining the party allegiance that they inherited
from their parents. Conservatives who were raised by Democratic or independent parents have
moved dramatically toward the Republican Party.92 Mason found that the increasing alignment of
partisan and ideological identities in U.S. society intensifies partisan bias and anger toward the
outgroup party. She argues that the result of this process is a nation that agrees on many things
but is bitterly divided nonetheless.93
To summarize, perceived credibility of new information is influenced by prior beliefs and
ideologies, and this effect is exacerbated by the use of intense or opinionated language (such as
language used to express binary legal judgments). In the U.S., liberals and conservatives, who hold
different sets of factual beliefs, respond differently to new information, based on its consistency
with their priors.
Based on this theoretical background, this study challenges the fundamental assumption
underlying international fact-finding: that more information means more shared knowledge; and
that legal fact-finding reports resolve controversies concerning war crimes and other violations of
international law. Specifically, I argue that we should expect fact-finding reports concerning war
crimes to be trusted by some social sub-groups, while rejected by others, depending on the report’s
judgment (whether it incriminates or absolves the relevant society from responsibility). For
example, in the U.S., I expect that individuals holding conservative ideological beliefs (who are
89 Id. at 958. 90 Alan Abramowitz, THE DISAPPEARING CENTER: ENGAGED CITIZENS, POLARIZATION, AND AMERICAN
DEMOCRACY 37 (2010). 91 Lilliana Mason, ‘I Disrespectfully Agree’: The Differential Effects of Partisan Sorting on Social and Issue
Polarization, 59 AM. J. POL. SCI. 128, 128 (2015). 92 Alan I. Abramowitz & Kyle L. Saunders, Ideological Realignment in the US Electorate, 60 J. POL. 634, 649
(1998). 93 Mason, supra note 88, at 142.
Krebs The Legalization of Truth
25
less critical of the U.S. military) will be more likely to discredit information that incriminates U.S.
soldiers for committing war crimes in Afghanistan than those holding liberal views (who are more
critical of the U.S. military). Similarly, individuals who hold conservative ideological beliefs will
be more likely to trust information absolving the U.S. from responsibility for war crimes than those
holding liberal ideological beliefs.94
(b) Emotional biases and receptiveness to fact-finding reports
Different legal terms have a distinct impact on people’s attitudes and beliefs.95 Research
on ‘hot cognition’ – cognition colored by feeling – suggests that cognitive and affective
determinants often work together to produce people’s judgments of what they think is just or unjust
and right or wrong.96 Words are agents of emotional judgment-making, as they automatically
activate affective, as well as semantic, associations.97 Lodge and Taber extended the affective and
semantic priming research into the political domain,98 and Morris et al. found that the emotional
evaluation of a political issue is stored with the concept.99 Later, when thinking about or called on
to evaluate this issue, the affective charge is spontaneously activated from long-term memory and
colors all subsequent cognitive and evaluative judgments about that issue.100 ‘Genocide,’ ‘crimes
against humanity,’ and ‘war crimes’ are all charged political terms which trigger negative
94 The study will therefore test the following hypothesis: If judgment=incriminating, Report Fairness(liberals)>50
and Report Fairness(conservatives)<50; If judgment=absolving, Report Fairness(liberals)<50 and Report
Fairness(conservatives)>50. 95 For example, in a study from 1995, McCaffery, Kahneman, and Spitzer demonstrated how different legal
framings of jury instructions led to substantive difference in compensation awards. Edward J. McCaffery, Daniel J.
Kahneman & Matthew L. Spitzer, Framing the Jury: Cognitive Perspectives on Pain and Suffering Awards, 81 VA.
L. REV. 1341, 1344 (1995). In a later study, focused on the framing effects of two different legal doctrines – strict
liability versus negligence – Cupp and Polage found that jurors are more likely to award damages if the information
they receive is presented using a negligence framing (a moral failing by the manufacturer to act reasonably) than if
the information is presented to them using a strict liability framing (a ‘cold’ legal doctrine that allows liability
regardless of blameworthiness). Richard L. Cupp Jr. & Danielle Polage, The Rhetoric of Strict Products Liability
Versus Negligence: An Empirical Analysis, 77 N.Y.U. L. REV. 874, 900 (2002). 96 Kees Van Den Bos, Hot Cognition and Social Justice Judgments: The Combined Influence of Cognitive and
Affective Factors on the Justice Judgment Process, in David De Cremer, ed., ADVANCES IN THE PSYCHOLOGY OF
JUSTICE AND AFFECT 59, 61 (2007). For a general introduction on hot cognition, see: ZIVA KUNDA, SOCIAL
COGNITION: MAKING SENSE OF PEOPLE (1999); PAUL THAGARD, HOT THOUGHT: MECHANISMS AND APPLICATIONS
OF EMOTIONAL COGNITION (2008). 97 Milton Lodge & Charles S. Taber, The Automaticity of Affect for Political Leaders, Groups, and Issues: An
Experimental Test of the Hot Cognition Hypothesis, 26 POL. PSYCHOL. 455 (2005). 98Id. 99 James P. Morris et al., Activation of Political Attitudes: A Psychophysiological Examination of the Hot Cognition
Hypothesis, 24 POL. PSYCHOL. 727, 741 (2003). 100 Id. at 727-28.
Krebs The Legalization of Truth
26
associations. They are commonly referred to by the media and are thus stored, together with their
negative emotional evaluation, in the long-term memory of individuals. When these terms are
invoked by fact-finding missions, they might trigger a spontaneous activation of these negative
emotional evaluations. For international lawyers, naming these violations is necessary to
accurately analyze events which might – or might not – violate international law, as defined by the
relevant legal framework. For society members, these legal terms – and their negative ‘hot’
political meaning –– may trigger negative emotional bias by society members in the shamed
communities, reactions that may frustrate attempts to introduce credible new information.
As previously explained, evaluation of information in social settings is influenced by a
complex set of beliefs, ideologies, and identities.101 People appear unable to break free of their
prior sentiments when evaluating arguments on political issues, even when they are motivated to
be impartial.102 While the previous section analyzed cognitive bias based on pre-existing
ideological commitments, this section focuses on emotional bias coupled with national identity
sentiment.
Identity is an individual's sense of self.103 It includes social identity, which captures her
knowledge that she belongs to certain social groups, together with some emotional significance to
her of this group membership.104 To enhance their social identities, individuals tend to behave in
ways that make their own group acquire positive distinctiveness in comparison to other groups.105
By the use of symbols such as flags, anthems, uniforms, monuments, and ceremonies, the society
members are reminded of their common heritage and cultural kinship and their sense of belonging
is enhanced.106 When beliefs regarding the national identity or national narrative begin to be
invalidated by incoming information, the individual experiences threat,107 and responds in a
101 Maoz et al., supra note 73, at 543. 102 Lodge & Taber, supra note 94, at 456. 103 CHARLES TAYLOR, SOURCES OF THE SELF: THE MAKING OF THE MODERN IDENTITY 3 (1989). 104 John C. Turner, Towards a Cognitive Redefinition of the Social Group, in SOCIAL IDENTITY AND INTERGROUP
RELATIONS 15, 18 (Henri Tajfel ed., 1982). 105 Ed Cairns, Intergroup Conflict in Northern Ireland, in SOCIAL IDENTITY AND INTERGROUP RELATIONS, supra
note 101, at 278; SAMUEL P. HUNTINGTON, WHO ARE WE? THE CHALLENGES TO AMERICA'S NATIONAL IDENTITY 25
(2004). 106 ANTHONY D. SMITH, NATIONAL IDENTITY 17 (1991); see also BENEDICT ANDERSON, IMAGINED COMMUNITIES
104 (2006); ERNEST GELLNER, NATIONS AND NATIONALISM (1983). 107 Kelly terms such beliefs 'core constructs.' GEORGE KELLY, A PSYCHOLOGY OF PERSONAL CONSTRUCTS (1955).
Krebs The Legalization of Truth
27
protective manner.108 Therefore, it is likely that the new invalidating information will be rejected
or distorted in order to fit the national narrative.109 As a shared national identity unites both liberals
and conservatives in the U.S., we should expect subjects to respond to threatening information
concerning their national identity in a nonpartisan manner. Unlike many other areas in American
politics, in which people tend to respond based on partisan affiliation, research shows that a
‘national interest’ framing of a topic prevents partisan divide. For example, Levendusky and
Horowitz demonstrated that presidential decisions which are framed in terms of national interest
influence the public to evaluate the president’s actions in a nonpartisan manner, wanting to support
the nation.110
Therefore, we should expect that condemning one’s nation using ‘hot’ political terms such
as war crimes, which have strong negative associations attached to them, may backfire.
Specifically, I argue that we should expect the ‘hot’ war crimes terminology concerning U.S.
military actions to trigger a defensive reaction that decreases the perceived credibility of the
information among U.S. nationals, regardless of their political ideology, in comparison to the
‘cold’ violation terminology. Even moderates, who are not expected to equally trust both absolving
and incriminating judgments (as they do not express strong ideological commitments concerning
the military), are predicted to respond negatively to the invocation of ‘war crimes’ terminology.111
Confirmation bias and cognitive consistency theories explain why liberals and conservatives
respond in opposite ways to new information, based on their pre-existing beliefs and ideological
commitments. However, hot cognition theory, together with threatened identity dynamics, explain
why a ‘war crimes’ framing negatively influences the perceived fairness and credibility of fact-
finding reports among U.S. nationals uniformly.
108 Northrup, supra note 72, at 66. 109 Id., at 69-70. Jervis defines this process as the "assimilation of information to pre-existing beliefs" while
implementing it to decision-making process in international relations. ROBERT JERVIS, PERCEPTION AND
MISPERCEPTION IN INTERNATIONAL POLITICS 143 (1976). 110 Matthew S. Levendusky & Michael C. Horowitz, When Backing Down Is the Right Decision: Partisanship, New
Information, and Audience Costs, 74 J. POL. 323-38 (2012). 111 The study will therefore test the following hypothesis: report’s perceived fairness hypothesized main effect: IJ >
WC; Liberals: IJ > WC > AJ; Conservatives: AJ > IJ < WC (IJ=incriminating judgment; AJ=absolving judgment
WC=war crimes).
Krebs The Legalization of Truth
28
(c) Judging truth by its cover: institutional legitimacy and receptiveness to fact-finding reports
Can acceptance of threatening information, or information that is inconsistent with prior
beliefs, be increased by deference to trusted legal institutions? Empirical research on source
credibility began in the twentieth century, and focused on the impact of source credibility on
attitude change and persuasion.112 Communications and social psychology literature on source
credibility has found evidence demonstrating that different perceptions of source credibility
differentially affect message evaluations.113 Hovland and Weiss defined source credibility in terms
of expertise and trustworthiness,114 and suggested that credibility dimensions will differ depending
upon the type of source being evaluated and the context in which the evaluation occurs.115
Following their studies, many researchers found evidence for both source expertise and source
trustworthiness as important dimensions influencing perceived credibility.116 Additionally, it was
established that positive regard for and similarity to the source influence audience perceptions of
trustworthiness and expertise.117 While the majority of these studies focused on individuals giving
a speech, recent studies tested the credibility of organizational sources, finding similar results.118
Nonetheless, while source credibility is an important factor in studying message evaluations, the
majority of the research suggests that the effects of source credibility are situational, and that
various factors have mediating effects on the impact of source credibility on message evaluations
(including factors relating to the message content, receivers’ characteristics, and a person’s initial
attitude toward the advocated position).119
Another dimension influencing message evaluations, stemming from a different theoretical
framework, is the perceived fairness of the process producing the relevant information. Procedural
justice studies have examined the impact of perceived fairness of the legal process on people’s
acceptance of their judgments. In their book Trust in the Law, Taylor and Huo find that acceptance
112 C.I. Hovland & W. Weiss, The Influence of Source Credibility on Communication Effectiveness, 15 PUB.
OPINION Q. 635 (1951). 113 Eric Haley, Exploring the Construct of Organization as Source: Consumers' Understandings of Organizational
Sponsorship of Advocacy Advertising, 25 J. ADVERTISING 19 (1996). 114 Hovland & Weiss, supra note 112, at 635-650. 115 Id. at 635. 116 For a meta-analysis surveying the relevant literature, see Metzger, supra note 67; see also Pornpitakpan, supra
note 55, at 247. 117 Metzger, supra note 67, at 298. 118 Id. at 299; Haley, supra note 113. 119 Haley, supra note 113.
Krebs The Legalization of Truth
29
of legal decisions increases when people regard the agents of the legal system as acting in a way
they perceive to be fair and trustworthy.120
Based on this theoretical literature, I suggest that discrepant or threatening messages will
be evaluated more positively coming from a trusted institution than from an untrusted one. As
public opinion polls show, conservatives in the U.S. are more likely to have favorable opinions of
U.S. military courts than liberals, and liberals are more likely to have favorable opinions of the
U.N. than conservatives. Accordingly, I argue that conservatives will be more likely to trust
counter-attitudinal information (reports incriminating U.S. soldiers for killing Afghan civilians),
if that information is conveyed by a U.S. military court rather than a U.N. fact-finding mission.
Similarly, I argue liberals will be more likely to trust counter-attitudinal information (absolving
fact-finding reports) coming from the U.N. rather than a U.S. Military Court.121
3. Challenging the Mobilization Assumption: Legal Truth Does Not Affect Attitudes
The previous section challenged the assumption that fact-finding reports concerning war
crimes resolve factual controversies concerning what happened (the credibility assumption). I
hypothesized that that the perceived credibility of a fact-finding report is determined by the
consistency of the report’s legal judgment with ideological commitments and prior beliefs, as well
as by the concrete legal terminology used to describe the events. The following section focuses on
the impact of legal frames on social attitudes, and challenges the assumption that exposure to
incriminating legal reports motivates in-group condemnation of the perpetrators. It argues, that
fact-finding reports face complex social dynamics which frustrate or diminish their ability to
meaningfully influence attitudes on accountability measures concerning war crimes.
(a) Naming, shaming and legal framing
Does incriminating legal framing mobilize people to support penalizing their compatriots
for committing war crimes against their enemies? Does international condemnation of the U.S.,
for example, for committing war crimes in Afghanistan, instigate a public outcry for punishing the
offenders? Many international and domestic advocacy organizations assume that it does. Their
120 TOM R. TYLER & YUEN J. HUO, TRUST IN THE LAW: ENCOURAGING PUBLIC COOPERATION WITH THE POLICE AND
COURTS 7 (2002). 121 Therefore, the study tests the following hypotheses: Liberals: if judgment=absolving (AJ), then fairness(UN) >
fairness(MC); Conservatives: if judgment=incriminating(IJ), then fairness(MC) > fairness(UN).
Krebs The Legalization of Truth
30
model is simple: incriminating fact-finding reports inform the public that crimes were committed,
and the public forms attitudes on sanctioning based on these findings. This article suggests a
different model. It argues that fact-finding reports have little impact on social attitudes concerning
accountability due to several psychological mechanisms.
First, some individuals are likely to reject the information altogether due to cognitive
inconsistency, based on their pre-existing beliefs and attitudes. Social-psychology literature has
demonstrated time and again that exposure to credible evidence is not enough to change people’s
attitudes on some issues. Mounting scientific evidence on the benefits of vaccinations, the
ineffectiveness of the death penalty, or the existence of global warming fail to change the attitudes
of those who oppose vaccinations, support the death penalty, or deny that global warming is
happening.122
Second, while some deny the factual evidence altogether, others accept the facts but
interpret them in a way that validates their pre-existing attitudes.123 For example, both Democrats
and Republicans in the U.S. accepted the fact that weapons of mass destruction were not found in
Iraq, but gave this fact a different meaning: for Democrats, the failure to find weapons of mass
destruction in Iraq meant that such weapons never existed, while Republicans inferred that Iraq
had moved, destroyed, or hidden the weapons.124
Third, others may accept the facts and agree on their interpretation, but disagree on the
solution or policy to address the issue. Acceptance of a set of facts does not necessarily mean
agreement on the appropriate course of action. For example, Campbell and Kay demonstrated that
Republicans’ increased skepticism toward environmental sciences may be partly attributable to a
conflict between specific ideological values and the most popularly discussed environmental
solutions.125 Interestingly, Republicans tripled their acceptance of scientific evidence on global
122 Brendan Nyhan et al., Effective Messages in Vaccine Promotion: A Randomized Trial, 133 PEDIATRICS e835-42
(2014); Donald Braman et al., Modeling Facts, Culture, and Cognition in the Gun Debate, 18 SOC. JUST. RES. 283
(2005); Dan M. Kahan et al., The Polarizing Impact of Science Literacy and Numeracy on Perceived Climate
Change Risks, 2 NATURE CLIMATE CHANGE 732-35 (2012); Lord, Ross & Lepper, supra note 77.
123 Gaines et al., supra note 80; see also George Loewenstein & Don A. Moore, When Ignorance Is Bliss:
Information Exchange and Inefficiency in Bargaining, 33 J. LEGAL STUD. 37 (2004). 124 Gains et al., supra note 80. 125 Troy H. Campbell & Aaron C. Kay, Solution Aversion: On the Relation Between Ideology and Motivated
Disbelief, 107 J. PERSONALITY & SOC. PSYCHOL. 809 (2014).
Krebs The Legalization of Truth
31
warming when it was coupled with a free-market-friendly solution rather than a government-
regulation solution.126
Similarly, we should not expect that attitudes on sanctioning soldiers who allegedly
violated the laws of war will change simply because a fact-finding report determines that they
should. Literature on crime and punishment demonstrates that sanctioning attitudes – just like
attitudes on gun control or global warming – are primarily determined by political and ideological
values and not by factual evidence concerning the circumstances of individual cases.127 As liberals
tend to be more critical of the U.S. military than conservatives, liberals are expected to be more
likely to condemn U.S. soldiers for misconduct on the battlefield. However, being critical of the
military and accepting incriminating evidence does not necessarily translate into unequivocal
support for prosecuting and punishing the soldiers. Some may believe that criminal prosecution is
too harsh, some may feel sympathy for the soldiers, some may find it sufficient to simply
acknowledge the wrongdoing, and some may assign blame to commanders and political leaders
rather than the soldiers on the ground, and argue that those higher up the chain of command should
bear responsibility. Therefore, even among those who find the incriminating judgment credible,
we should expect variation in the willingness to impose any penalties on soldiers. In the end, the
legal judgment of a fact-finding report in itself is not sufficient to influence attitudes and mobilize
domestic sanctioning of in-group offenders. I therefore predict that only a minority of U.S.
nationals will support penalizing U.S. soldiers after being exposed to a fact-finding report accusing
them of committing war crimes, and that the legal judgment – whether incriminating or absolving
– will not determine the willingness to penalize the soldiers.128
(b) Are moral judgments more powerful than legal judgments?
If, indeed, legal discourse is ineffective in mobilizing domestic condemnation of in-group
offenders, is there a better way to mobilize domestic support for accountability?129
126 Campbell and Kay, id. 127 Tom R. Tyler & Renee Weber, Support for Death Penalty: Instrumental Response to Crime, or Symbolic
Attitude?, 17 LAW & SOC’Y REV. 21, 43 (1982). 128 Therefore, this article tests the following hypotheses: If legal judgment=incriminating: Prosecute(0-100)<50,
Condemn(0-100)<50, Compensate(0-100)<50; prosecute(IJ)=Prosecute(AJ). 129 In Lowering the Bar, Marc Galanter documented the social perception of law and lawyers in the U.S. He found
that law and lawyers have typically been associated with negative traits such as greed, aggressiveness, indifference
to justice, manipulation, and opportunism. “Lawyers corrupt discourse by promoting needless complexity,
Krebs The Legalization of Truth
32
In her article Words and the Door to the Land of Change, Martha Minow explores the
potential power of words to effectuate social change. She observes that judges use a certain
language – like that of a remote clinical evaluation – and adopt neutral and obscure phrases like
“abuse,” “multiple bruises and abrasions,” and “suspicious injuries,” which undermine the sense
of personal connection and individual obligation to effectuate change.130 She wonders “whether
words by lawyers and judges differ from words by journalists, and whether the more intimate and
yet more widely accessible languages of literature and popular music lyrics may change minds and
prompt actions.”131 Minow suggests that different forms of expression solicit different responses,
and argues that “finding languages to persuade judges, to empower victims, and to mobilize
onlookers present linked yet distinct difficulties.”132 Drawing on Minow’s argument, I suggest that
the use of legal discourse to describe facts might be counter-productive. To test this argument, I
compare the impact of legal and moral condemnations on willingness to sanction in-group
offenders and to compensate out-group victims.133
4. The Legalization of Truth: Problems and Solutions
This article argues that the most common organizational structure of international fact-
finding – a legal-discourse-based structure, lacking enforcement capacity – is based on a flawed
understanding of the relevant socio-psychological processes and dynamics. The binary legal
judgment or outcome of the fact-finding report triggers ideological-based belief polarization; the
adoption of ‘hot’ legal terminology reduces the perceived fairness of the report; and the legal frame
is less effective than a moral frame in influencing attitudes. Overall, the legal discourse is
ineffective in creating a shared history, disseminating an authoritative account of contested facts,
and mobilizing domestic attitudes to condemn in-group offenders. When faced with unsettling
facts, various cognitive and emotional biases prevent individuals from accepting these facts at face
value and modifying their attitudes accordingly. Some individuals are likely to reject the
mystifying matters by jargon and formalities, robbing life’s dealings of their moral sense by recasting them in legal
abstractions, and offending common sense by causistry that makes black appear white and vice versa.” MARC
GALANTER, LOWERING THE BAR: LAWYER JOKES AND LEGAL CULTURE 17 (2005). 130 Martha Minow, Words and the Door to the Land of Change: Law, Language, and Family Violence, 43 VAND. L.
REV. 1665, 1666 (1990). 131 Id. at 1673, 1685. 132 Id. at 1665-1666. 133 Specifically, I will test the following hypothesis: Prosecute(IMJ) > Prosecute(ILJ); Condemn(IMJ) >
Condemn(ILJ); Compensate(IMJ) > Compensate(ILJ).
Krebs The Legalization of Truth
33
information based on its inconsistency with their prior beliefs; some are likely to reinterpret the
report to fit their ideological commitments; and others may accept the facts but reject their legal
implications. Therefore, the core assumptions underlying international fact-finding efforts – that
they contribute to resolving factual disputes and mobilizing domestic sanctioning – should be
questioned.
To test these assumptions and to more directly study the domestic effects of fact-finding
reports, while mitigating problems of selection effects, I designed and fielded a pair of survey-
experiments for the U.S. public.134 To operationalize my hypotheses concerning the impact of the
legalization of truth on both the perceived credibility of the fact-finding report and the willingness
to sanction in-group offenders, I focused on three main aspects of legal discourse: first, the binary
legal judgment or outcome, comparing an incriminating judgment with an absolving one. Second,
the legal terminology, comparing cold ‘violation’ language with hot ‘war crime’ language. Third,
the legal frame itself, comparing legal and moral frames describing the exact same findings and
outcomes. Experiment I (conducted in April 2014) measures the perceived credibility of a fact-
finding report, given three legal judgments: two incriminating legal judgments (“violating
international law” and “committing war crimes”) and one absolving legal judgment (“not violating
international law”); and three institutions rendering these reports: the United Nations, a U.S.
military court and a U.S. federal court. Experiment II (conducted in December 2013), measures
the persuasive power of moral, as opposed to legal, framing of the report’s findings, and the
respective impact on attitudes concerning accountability measures.
Random assignment increases the likelihood that treatment and control groups are similar
to each other on average across all observed and unobserved factors. By comparing differences in
responses between groups, an experimental design can estimate with greater certainty the effect of
legal truth on factual beliefs and attitudes regarding domestic sanctioning of in-group offenders.135
Unlike observational date (such as national surveys), the experimental approach provides us with
a direct measure to assess the efficacy of fact-finding reports, depending on their main
characteristics, including the legal discourse, the final judgment, and the institution establishing
134 These experiments were back-translated and adjusted for use in other countries, including Israel and Canada, to
provide comparative results. 135 Wallace, supra note 19, at 116.
Krebs The Legalization of Truth
34
the mission. The limitations of experimental methods include, of course, their external validity –
the degree to which results are generalizable to broader phenomena of interest. Experiments also
reduce scenarios to a few key variables, often implemented over a short time period, compared to
the complex and fluid nature of everyday relations. Subjects may act differently in laboratory
settings than in real life conditions. I tried to alleviate these issues by conducting surveys at
different points in time; using nationally representative samples; and providing vignettes that
closely resemble real-world situations. Additionally, in two separate studies I test the
generalizability of these findings, first, to other societies, by fielding similar experiments in other
countries, including Canada and Israel; and, second, to domestic law, testing this theoretical
framework on the current controversy in the U.S. concerning police violence against African
Americans and the ‘Black Lives Matter’ movement. Taking these limitations in mind, the
experimental evidence that follows provides a basis for further inquiry.
C. EXPERIMENT I: THE EFFECTS OF LEGAL TRUTH ON THE PERCEIVED CREDIBILITY OF FACTS
Experiment I was fielded in April 2014 as part of the 2014 American National Opinion
Survey (ANOS), an original internet survey of U.S. citizens carried out by YouGov. YouGov uses
sampling and matching techniques to generate a sample that approximates the demographic
composition of the adult U.S. population.136
1. Design
Two thousand respondents were randomly assigned to one of three legal judgments
(judgment condition: ‘cold’ incriminating judgment (‘violating international law’ (IJ)), ‘hot’
incriminating judgment (‘committing war crimes’ (WC)), and ‘cold’ absolving judgment (‘not
violating international law’ (AJ)), and to one of three institutions (United Nations (UN), United
States Military Court (MC) and United States Federal Court (FC).
Respondents were told that they would be asked questions about a military operation in
Afghanistan which happened a few months ago. They were also told that since the facts of the case
were unclear (especially how many people died, and whether the casualties were Al-Qaeda
136 See Appendix I.2 for further information about the construction of the ANOS sample; see Table 1 for descriptive
statistics of the sample.
Krebs The Legalization of Truth
35
members or innocent civilians), an investigation was conducted by one of the three institutions
mentioned above. Respondents were then presented with the investigation’s executive summary,
which included two parts: a ‘summary of facts’ and a ‘summary of the legal judgment.’ Both parts
were designed to look like a formal document, including the institution’s logo, the document’s
serial number, and the date.
The summary of facts described an event that happened recently in Afghanistan, in which
a Taliban fighter killed a U.S. Marine with an explosive device. Three Marines, who witnessed the
explosion, chased the attacker into a nearby marketplace, shooting their guns repeatedly.
Eventually, they were able to kill the attacker, but four unarmed Afghan bystanders were also
killed by the Marines’ gunfire. The wording of the ‘summary of facts’ was identical for all three
conditions, and the only differences were the logo and name of the institution producing the report.
After reading the factual summary of the events, respondents were presented with the legal
judgment. First, the report stated the relevant legal standard, according to which soldiers are
obliged to exercise reasonable care to protect civilians during armed conflicts. Second, the report
applied this standard to the facts of the case, concluding that the U.S. Marines violated international
law, committed war crimes, or did not violate international law.
After reading the report’s summary, respondents were then asked three questions on the
credibility of the report (its accuracy, objectivity, and fairness). Additionally, the survey included
demographic, political, social, and economic measurements, including gender, education,
ideology, political identification, religiousness, interest in news, and financial situation.137 The
three judgment-condition structure implies a factorial design with three experimental groups. The
full text for this and the subsequent survey instrument are provided in Appendix I.3.
This design was very carefully constructed to be as realistic as possible without inserting
any bias for or against sanctioning the soldiers. First, I chose a scenario from the war in
Afghanistan because this conflict is less politicized in American public opinion than the war in
Iraq; moreover, it adds to the credibility of the scenario, as this conflict is ongoing, with American
troops still deployed in Afghanistan. Second, I calibrated the scale or severity of the event to allow
137 For a full description of the experimental treatments and the surveys questions, see Appendix I.3. For a full list of
the measurements used, see Appendix I.2.
Krebs The Legalization of Truth
36
for credible variation in the legal judgment and decrease social desirability bias. Survey questions
asking about sensitive issues often generate socially motivated misreporting and distort survey
estimates.138 In other words, due to self-presentation concerns, survey respondents underreport
socially undesirable activities and overreport socially desirable ones. In this case, designing an
experimental scenario in which U.S. soldiers are responsible for large-scale war crimes and crimes
against humanity could have motivated respondents to conceal their true attitudes, finding it
socially undesirable to support soldiers who are responsible for mass killings. Adopting an extreme
scenario could have therefore triggered social desirability bias which would have generated
estimate errors concerning willingness to condemn and prosecute soldiers who violated
international law. Additionally, a mass killing scenario would have decreased the credibility of the
vignette, as an absolving judgment may have seemed incongruous following a description of a
large-scale massacre. To be able to credibly vary the legal judgment based on the same facts, it
was necessary to design a smaller-scale incident. Third, typically, individuals are exposed to this
type of information through the media. However, I decided against using media coverage of the
report, and instead presented participants with a direct ‘quote’ of the executive summary of the
report, in order to observe the reaction to the report and its source, rather than to the media outlet
reporting about it. This design was carefully constructed to capture real-life reactions to fact-
finding reports, and was pretested on 600 Amazon M-Turk users to test for manipulation checks
and believability of the vignette, and to allow respondents to comment, express, and explain their
reactions.
By design, random assignment improves the chances experimental groups differ only
with respect to the treatment. As expected, tests indicate groups assigned to the three treatment
groups were comparable across all observed characteristics.139
138 Ivar Krumpal, Determinants of social desirability bias in sensitive surveys: a literature review, 47(4) QUALITY &
QUANTITY 2025, 2035-2036 (2013).
139 I conducted an extensive series of balance tests comparing the distribution across treatment groups for all
available baseline covariates that likely affect attitudes concerning war crimes, including gender, age, education,
income, religion, and race.
Krebs The Legalization of Truth
37
2. Measures
Report’s Credibility. The literature suggests that the perceived credibility of information has
several dimensions, including accuracy, objectivity, and fairness.140 On the basis of these previous
studies, I used three items to assess the perceived credibility of the report, its factual findings, and
its final conclusion/judgment (“Do you think that the facts summarized by the investigators are:
Completely inaccurate… Completely accurate?”; “Do you think that the report is: Completely
unfair… Completely fair?”; “Do you think that the judgment is: Completely biased… Completely
objective?”). Respondents indicated their attitudes on a 6-point scale; these data were later recoded
to range from 0 (completely unfair/inaccurate) to 100 (completely fair/accurate). The same
response scale was used for all other measures in Experiment I unless otherwise noted. As these
three items were highly correlated, for simplicity, the analysis presents the results for the fairness
variable, which captured the overall reaction to the report as a whole.
Institution’s favorability. Responders were asked for their opinion on the relevant institutions (the
United Nations, U.S. federal courts and U.S. military courts). Participants indicated their attitudes
on a 4-point scale, from ‘very favorable’ to ‘very unfavorable’; these data were later recoded to
range from 0 (very unfavorable) to 100 (very favorable).
Political ideology. Respondents rated their political ideology on a 5-point scale (1=“very liberal”;
3=“moderate”; 5=“very conservative”), and indicated the level of their political party identification
on a 7-point scale (1=“strong democrat”; 4=“independent”; 7=“strong Republican”).
Control Variables. Because of randomization, complex statistical models involving a battery of
control variables are unnecessary for obtaining valid inferences regarding the effect of
international fact-finding. Nonetheless, to increase confidence in any empirical findings, I
conducted a set of regressions which controlled for demographic variables including gender,
140 Andrew J. Flanagin & Miriam J. Metzger, Perceptions of Internet Information Credibility, 77 JOURNALISM &
MASS COMM. Q. 515 (2000); Cecilie Gaziano & Kristin McGrath, Measuring the Concept of Credibility, 63
JOURNALISM Q. 451-62 (1986); Philip Meyer, Defining and Measuring Credibility of Newspapers: Developing an
Index, 65 JOURNALISM Q. 567-74 (1988); John Newhagen & Clifford Nass, Differential Criteria for Evaluating
Credibility of Newspapers and TV News, 66 JOURNALISM Q. 277-84 (1989); Tony Rimmer and David Weaver,
Different Questions, Different Answers? Media Use and Credibility, 64(1) JOURNALISM & MASS COMM. Q. 28
(1987); Eugene F. Shaw, Media Credibility: Taking the Measure of a Measure, 50(2) JOURNALISM & MASS COMM.
Q. 306 (1973); Mark Douglas West, Validating a Scale for the Measurement of Credibility: A Covariance Structure
Modeling Approach, 71 JOURNALISM Q. 159-68 (1994).
Krebs The Legalization of Truth
38
education, income, and age. The effect of legal judgment became even more robust controlling for
various demographic variables.
3. Results
(a) Legal discourse fails to resolve factual controversies
The main effect of the legal judgment (whether incriminating or absolving) was not
significant across subjects (t(1,312)=1.02, p=.3). However, as shown in Figure 1, the effect of the
legal judgment was significantly different for liberals and conservatives: liberal subjects showed
a preference for incriminating legal judgment (mean difference t(323)=-8.26, p=.000), whereas
conservatives showed a preference for absolving legal judgment (mean difference t(447)=7.8,
p=.000). Moderates did not show any preference (t(421)=.2, p=.84). This means there is a
distinction between what liberals and conservatives in the U.S. believe to be true: conservatives
are especially skeptical of incriminating reports, particularly when compared with liberals, and
liberals are skeptical of absolving reports, particularly when compared with conservatives. Figure
2 shows that, in contrast to individuals with pre-existing ideological convictions, individuals who
self-report as moderates attribute similar levels of fairness to both incriminating and absolving
reports, which are lower than the perceived credibility liberals and conservatives attribute to
information consistent with their ideological commitments, but higher than the perceived
credibility liberals and conservatives attribute to information inconsistent with their priors.141
Next, I conducted a linear regression of perceived fairness of the report (the dependent
variable) on the legal judgment (dichotomous independent variable), ideology, and their
interaction. As shown in Figure 1, the model revealed the predicted interactive pattern b=-14.6,
SE=1.2, t(1,193)=-11.6, p<0.001. This confirms that different social groups respond in different
ways to new evidence concerning war crimes.
141 See appendix I.4, table 3, for summary of treatment effects.
Krebs The Legalization of Truth
39
Figure 2. Report Fairness, by Judgment Condition and Ideology
Figure presents the mean values of the Report Fairness, by ideology. 95% CI’s shown.
(b) The invocation of “war crimes” is counterproductive
I tested the impact of ‘war crimes’ on the perceived fairness of the report. As expected, the
main effect for war crimes was significant and negative: t(1320)=-2.75, p<.01. In other words,
invoking a ‘war crimes’ terminology decreases the perceived fairness of the report compared with
a ‘violation’ terminology. While statistically significant, this effect is small (Cohen’s d=0.15) and
accounts for about 5% change in the perceived fairness of the report. The negative effect of war
crimes terminology on perceived fairness is not moderated by ideology. There was a significant
difference between perceived fairness of war crimes and violation for conservatives (t(410)=-2.3,
p=.01) and moderates (t(442)=-1.9, p=.05), but although the direction was as expected with regard
to liberals (means difference -3.6), this difference was not significant (t(352)=-1.44, p=.14). Figure
3 demonstrates the main effect of the legal judgment framing (“war crimes” versus “violation”) on
Krebs The Legalization of Truth
40
perceived fairness of the report, and Figure 4 shows the means difference of war crimes and
violations framings for liberals, moderates, and conservatives.
Figure 3. Figure 4. Report Fairness, Report Fairness, by Legal Terminology by Legal Terminology and Ideology
Figure presents the mean values of the Report Fairness, by Legal Terminology and Ideology. 95% CI’s shown.
(c) Legal judgment trumps an institution’s favorability
In general, U.S. nationals have a more favorable opinion of U.S. military courts than of the
United Nations. A majority of Americans (62%) have a favorable opinion of U.S. military courts,
while only about half of Americans (51%) have a favorable opinion of the United Nations.142
However, the level of support varies considerably by political ideology. A strong majority of
liberals in the U.S. (77%) have a favorable view of the U.N., while a smaller majority of moderates
(58%) agree. Meanwhile, support among conservatives trails at 24%. Conversely, a strong
majority of conservatives in the U.S. (76%) have a favorable opinion of U.S. military courts, while
a smaller majority of moderates (62%) and a minority of liberals (45%) agree. To demonstrate the
moderating effect of political ideology on institution favorability, I conducted a linear regression
of institution favorability on institution (UN/MC), ideology, and their interaction. The model
revealed the predicted interactive pattern, b=.6, SE=.04, t(1,093)=13.3, p<0.001. This confirms the
142 As shown in Figure 5, I found a main effect for Military Court on favorability: b=10.5, t(1,163)=5.7, p<.001).
Krebs The Legalization of Truth
41
findings of previous studies on trust in international institutions,143 which found that liberals
demonstrate a more favorable opinion of the U.N. compared with conservatives.
I hypothesized that these institution-favorability trends translate to an increase in the
perceived fairness of counter-attitudinal reports, when the counter-attitudinal report is issued by a
trusted institution. Namely, liberals will be more likely to trust absolving reports issued by the
U.N. than those issued by U.S. military courts, and that conservatives will be more likely to trust
incriminating reports issued by U.S. military courts rather than those produced by the U.N. This
hypothesis is further supported by the goals and agendas of both institutions: if the U.N. report
which was specifically designed to ‘end impunity’ finds the U.S. Marines innocent, it must be true;
and similarly, if the U.S. military court itself finds U.S. Marines guilty, then it must be so.
However, as shown in Figures 5 and 6, the results failed to reject the null hypotheses that the
identity of the institution has no effect on the perceived fairness of counter-attitudinal reports.144 I
found that both liberals and conservatives were slightly more willing to trust counter-attitudinal
reports issued by the U.S. military court, but both findings were statistically insignificant.
Figure 5. Figure 6. Report Fairness (Conservatives), Report Fairness (Liberals), by Legal Judgments and Institution by Legal Judgments and Institution
Figures presents the mean values of the Report Fairness, by Legal Judgment and Institution. 95% CI’s shown.
143 U.N. Retains Strong Global Image: Robust Support in America, Especially Among Democrats, PEW RESEARCH
CENTER (Sep. 2013), http://www.pewglobal.org/2013/09/17/united-nations-retains-strong-global-image/. 144 For conservatives presented with an incriminating report, the difference of means had the predicted direction, but
no statistical significance (t(136)=-1.18, p=0.23). For liberals presented with an absolving report, the difference of
means did not have the predicted direction and was also not statistically significant (t(101)=-1.3, p=0.19).
Krebs The Legalization of Truth
42
Nonetheless, I did find that the polarizing effect of legal judgment on report fairness (the
moderating effect of political ideology on report fairness given different legal judgments) was
mitigated when the report was issued by the U.S. military court. That is to say, given an
incriminating legal judgment, the difference of means for liberals and conservatives was 26.25
(Cohen’s d=1.06, large effect size) if the U.N. issued the report and only 17.83 (Cohen’s d=0.66,
medium effect size) if the report was issued by the U.S. military court. Similarly, given an
absolving legal judgment, the difference of means was -25.15 (Cohen’s d=0.9) if issued by the
U.N. and only -14.44 (Cohen’s d=0.5) if issued by the U.S. military court.145
D. EXPERIMENT II: THE EFFECT OF LEGAL AND MORAL FRAMINGS ON WILLINGNESS TO
SANCTION IN-GROUP OFFENDERS
Experiment II was fielded within the 2013 American National Opinion Survey (ANOS),
an original internet survey of U.S. citizens carried out by YouGov. One thousand respondents were
randomly assigned to the control group (N=185) or one of four treatment groups: Incriminating
Legal Judgment (N=199), Absolving Legal Judgment (N=206), Incriminating Moral Judgment
(N=213), and Absolving Moral Judgment (N=197).146
1. Design
The design was similar to that of Experiment I: subjects were told that they would be asked
questions about a military operation in Afghanistan which happened a few months ago. They were
also told that since the facts of the case were unclear (especially how many people died, and
whether the casualties were Al-Qaeda members or innocent civilians), a team of international
experts was asked to investigate. Treated subjects were then presented with the summary of the
investigators’ factual findings, followed by a summary of the investigators’ judgment (that the
investigation concluded that the Marines acted illegally, legally, immorally, or morally). Control-
group subjects were only presented with the summary of the factual findings (no
judgment/conclusion was provided).
145 See appendix I.4, table 3, for summary of treatment effects. 146 See Appendix I.2 for further information about the construction of the ANOS sample. See Table 2 for descriptive
statistics of the sample.
Krebs The Legalization of Truth
43
The summary of facts was identical to the one used in Experiment I, and was presented to
all participants. The summary of conclusions was similar to the one used in Experiment I, but in
this experiment I introduced a non-legal conclusion, based on a moral standard. The moral
conclusions were similar to the legal “violation” treatment used in Experiment I, but replaced the
term ‘international law’ with ‘universal moral standards’ and the words ‘legal standards’ and ‘legal
responsibility’ with ‘moral standard’ and ‘moral responsibility.’ Otherwise the experiment was
identical to Experiment I.
2. Measures
Willingness to sanction. Three items were used to assess the degree of willingness to
sanction in-group offenders. First, I measured the subjects’ opinion of the soldiers’ behavior, and
their willingness to condemn the soldiers’ behavior (“The U.S. Marines did the best they could
under the circumstances.”). This item was recoded as the reversed item (“The U.S. Marines did
not do the best they could under the circumstances.”). Second, I measured the subjects’ support
of prosecuting the soldiers (“The U.S. Marines should be prosecuted.”). Third, I measured the
subjects’ support of compensating the Afghan victims (“The US Government should compensate
the Afghan victims.”)
All other measures were identical to Experiment I.
3. Results
Before turning to discussion of the unique results of this experiment, I should mention that
findings on the interaction model of legal judgment and ideology on report fairness were replicated
in this experiment, thus providing further support for the findings of Experiment 1.
(a) The legalization of truth and attitude mobilization
To test the impact of legal judgments on attitude mobilization, I first conducted a linear
regression of willingness to condemn the soldiers (dependent variable) on legal judgment. The
model revealed the predicted pattern, b=-2.08, SE=3.12, t(403)=-.66, p>.5. I conducted another
linear regression of willingness to prosecute the soldiers on legal judgment, and found a similar
pattern: b=-1.01, SE=3.14, t(403)=-0.32, p>.7. In both cases, participants were just as unwilling to
condemn the soldiers’ behavior or to support prosecuting them when told that the soldiers violated
Krebs The Legalization of Truth
44
the law or did not violate the law. All four means represent an unchanged negative attitude on
condemning or prosecuting the soldiers. In short, the legal judgment did not influence attitudes or
mobilize support for accountability measures.147
Interestingly, the one variable that was affected by the legal judgment was willingness to
compensate the victims. However, respondents were more willing to compensate the victims when
told that the soldiers did not violate the law, rather than when they were told the soldiers did violate
the law. In other words, given an absolving legal judgment, participants did not show a negative
attitude on compensating the victims. Figure 7 summarizes these findings.
Figure 7. Willingness to Condemn, Prosecute, and Compensate, by Legal Judgment
Figure presents the mean values of Willingness to Condemn the Soldiers, Willingness to Prosecute the Soldiers, and Willingness to Compensate the Victims, by Legal Judgment. 95% CI’s shown.
Figures 8 and 9 reveal that, as expected, willingness to condemn or prosecute the soldiers
is largely dependent on political ideology, and only marginally affected (if at all) by the legal
judgment. In fact, the only effect of legal judgment that was statistically significant was the
147 See appendix I.4, table 4, for summary of treatment effects.
Krebs The Legalization of Truth
45
positive effect of the absolving legal judgment (in comparison to the incriminating legal judgment)
on willingness to compensate the victims.
Figure 8. Figure 9. Willingness to Prosecute, Willingness to Compensate, by Legal Judgment and Ideology by Legal Judgment and Ideology
Figures presents the mean values of Willingness to Prosecute the Soldiers and Willingness to Compensate the Victims, by Legal Judgment and Ideology. 95% CI’s shown.
Figure 10 summarizes the impact of incriminating legal reports on domestic beliefs and
attitudes in the U.S.: a small majority of participants (56%) found the report fair, and of those, only
about half (26% of all participants) supported prosecuting the perpetrators. Of the 44% of
participants that found the report to be unfair, almost all were unwilling to support prosecuting the
perpetrators. While I found both ideology and gender to be predictors of support in prosecuting in-
group perpetrators, more data should be collected to fully explain why a belief that soldiers
committed war crimes does not translate into willingness to sanction them.
Krebs The Legalization of Truth
46
Figure 10. Willingness to Prosecute, by Report’s Fairness (Incriminating Judgment Conditions)
Figure presents the percentage of individuals who indicated their support or opposition for prosecuting the soldiers, by acceptance or rejection of the Report Fairness. 95% CI’s shown.
(b) Are there more effective alternatives to legal truth?
The results indicate that, as expected, legal truth was not effective in mobilizing domestic
attitudes on war crimes. But can different language or framing of the events be more effective in
influencing such attitudes?
To test the hypothesis that moral framing may be more effective in mobilizing domestic
attitudes on war crimes, I conducted a linear regression of willingness to prosecute on
incriminating judgment framing (dichotomous variable, moral judgment =1). The model revealed
the predicted pattern, b=9.1, SE=3, t(411)=2.9, p=.003 (Cohen d=-0.29). This shows that
respondents who were told that the soldiers violated moral standards were significantly more
willing to prosecute the soldiers than those who were told that the soldiers violated legal standards.
I repeated the regression analysis with both condemnation and compensation variables, with
similar results: condemning the soldiers: b=7.3, SE=2.9, t(411)=2.4, p=.01; compensating the
victims: b=8.6, SE=3.31, t(411)=2.6, p<.01.
Krebs The Legalization of Truth
47
Figure 11. Willingness to Prosecute, by Judgment Condition
Figure presents the mean values of Willingness to Prosecute the Soldiers, by Experimental Condition (Legal Judgment and Framing). 95% CI’s shown.
Next, I tested whether these findings are moderated by ideology. As reported in table 4,
liberals are generally more willing to prosecute U.S. soldiers than moderates and conservatives.
However, the main effect of moral frame of incriminating judgment is not moderated by ideology,
and the incriminating moral frame treatment increased willingness to prosecute for liberals,
moderates and conservatives.148
148 See appendix I.4, table 4, for summary of treatment effects.
Krebs The Legalization of Truth
48
E. DISCUSSION AND THEORETICAL IMPLICATIONS
What happens to these reports when they are “released?” The resources that organizations devote to compiling all this information are not matched by attention to how reports are disseminated or what impact they might have on target audiences.149
This article sheds light on ‘what happens’ to international fact-finding reports when they
are released. It focuses on their domestic audience, and examines how different legal and moral
framings of incriminating and absolving judgments influence the perceived credibility of the
reports and the willingness to sanction in-group offenders. Its main findings are as follows:
1. More Information, More Bias
Experiment I demonstrated that fact-finding reports are trusted by some social sub-groups,
while rejected by others, depending on the report’s legal judgment (whether it incriminates or
absolves the relevant individuals or social groups from responsibility). Specifically, it
demonstrated that in the U.S., individuals who hold conservative ideological beliefs are more
likely than those who hold liberal views to discredit information that incriminates U.S. soldiers for
committing war crimes in Afghanistan. Similarly, it found that individuals who hold liberal
ideological beliefs are more likely than those holding conservative ideological beliefs to discredit
information absolving the U.S. from responsibility for war crimes. This finding was replicated in
Experiment II. While these findings are consistent with similar findings relating to other sources
of information (‘the hostile media bias’150) and other areas of factual disagreements (death
penalty,151 global warming152), it is nonetheless important to establish that legal judgments
conveyed by international or domestic institutions might trigger similar reactions with regard to
issues such as war crimes. Establishing that perceived credibility of fact-finding reports is largely
determined by political and ideological commitments has unique significance with regard to legal
institutions, which unlike the media, are authoritative in making factual determinations concerning
legal disputes.153 One possible explanation for this finding lies in the fundamental disagreements
149 Cohen, supra note 2, at 518. 150 Robert P. Vallone, Lee Ross & Mark R. Lepper, The Hostile Media Phenomenon: Biased Perception and
Perceptions of Media Bias in Coverage of the Beirut Massacre, 49 J. PERSONALITY & SOC. PSYCHOL. 577 (1985). 151 Lord, Ross & Lepper, supra note 77, at 2098. 152 Kahan et al., supra note 122, at 732. 153 Kahan, Hoffman & Barman, supra note 76.
Krebs The Legalization of Truth
49
on substantive questions of international law and their applicability to different contexts. In the
decentralized international legal system, the interpretation of international law raises many
unresolved controversies. Eyal Benvenisti argues that the content of international humanitarian
law (or law of armed conflict) depends on the identity of the interpreting body – whether it is a
government involved in transnational armed conflict, or an international organization.154 Adding
a layer of disputable legal conclusions to factual determinations exposes the factual description to
criticisms based on a disagreement with the legal conclusion, and shifts the attention from the facts
to their legal meaning. If the content of international law is subjective, as some argue, it is not at
all surprising that its legalized fact-finding reports are treated similarly to other sources of
information, that their perceived credibility is moderated by political ideology, and that instead of
a deliberative process of truth-seeking, as suggested by Thomas Risse,155 we witness a growing
polarization concerning both the facts and their meaning.
2. War Crimes Terminology Decreases Perceived Credibility of Fact-Finding Reports
The language and terms we choose influence how our message is being processed and
assimilated. Just as with any other organizing themes, legal terminology creates frames under
which information is being categorized and interpreted.156 McCaffery, Kahneman, and Spitzer
demonstrated how different legal terms used in jury instructions have led to substantive differences
in compensation awards.157 Cupp and Polage found that jurors are more likely to award damages
if the information they receive is presented using a negligence framing (a ‘hot’ legal terminology
which assigns responsibility based on blameworthiness) than if the information is presented to
them using a strict liability framing (a ‘cold’ legal terminology, assigning responsibility regardless
of blameworthiness).158
The results of Experiment I demonstrate that similarly, the choice of legal terminology,
and specifically, between ‘hot’ war crimes terminology and ‘cold’ violations terminology,
154 Eyal Benvenisti, The Legal Battle to Define the Law on Transnational Asymmetric Warfare, 20 DUKE J. COMP. &
INT'L L. 339, 358-59 (2009). 155 Thomas Risse, “Let's argue!”: Communicative Action in World Politics, 54 INT’L ORG. 1 (2000). 156 McCaffery, Kahenman & Spitzer, supra note 92, at 1353. See, also, Richard K. Sherwin, Law Frames: Historical
Truth and Narrative Necessity in a Criminal Case, 47(1) Stanford Law Review 39 (1994). 157 McCaffery, Kahneman & Spitzer, supra note 92, at 1340. 158 Richard L. Cupp Jr. & Danielle Polage, The Rhetoric of Strict Products Liability Versus Negligence: An
Empirical Analysis, 77 N.Y.U. L. REV. 874, 900 (2002).
Krebs The Legalization of Truth
50
influenced the perceived credibility of fact-finding reports. The war crimes terminology decreased
the perceived fairness of the report in comparison to the legal violation terminology. David
Scheffer, who served as the United States Ambassador-at-Large for War Crimes Issues at the U.N.,
suggested that legal terms such as ‘war crimes’, ‘crimes against humanity,’ and ‘genocide’ are so
tainted by political meanings that they should not be used by legal institutions at all.159 While some
Americans believe that the U.S. occasionally violates international law in the course of armed
conflicts, they are less willing to go so far as to characterize these violations as war crimes.
Interestingly, moderates, who did not hold strong priors and were just as likely to believe
incriminating and absolving reports, were less likely to believe the report when the war crimes
terminology was used. These findings suggest that several cognitive and emotional biases have
distinct impacts on individuals from different social groups when processing new information.
While cognitive consistency is at play when one confronts counter-attitudinal facts, hot cognition
and threatened identity dynamics become more pronounced when national narratives are
challenged.
As Scheffer correctly identified, when individuals think about ‘war crimes’ they attach a
certain political meaning to the phrase. In this case, as is often the case in real life, the legal
judgment of the case was debatable. In such a situation, while ‘war crimes’ might be the correct
legal term, non-lawyers might adopt a political, rather than legal, interpretation of this term, and
attach a different meaning to it, including concerning the gravity of events that might justify
invoking it. In the context of the U.S. war in Afghanistan, these findings suggest that the use of
war crimes terminology triggers emotional biases because of the salient negative political meaning
of this term.
3. Legal Judgment Trumps the Issuing Institution’s Favorability
In contrast to my initial hypothesis, the study was unable to detect significant effects of the
issuing institutions on perceived fairness of the reports. This result may be explained by the
importance and salience of the legal judgment effect. Given a counter-attitudinal legal judgment,
on such a ‘hot’ topic, the impact of source favorability was mitigated. Another alternative
explanation is that the institution treatment was too subtle and not salient enough to trigger attitude
159 David Scheffer, Genocide and Atrocity Crimes, 1 GENOCIDE STUD. & PREVENTION 229 (2006).
Krebs The Legalization of Truth
51
change. Lastly, perhaps neither institution was trusted enough to trigger attitude change. It could
be the case that to disseminate counter-attitudinal facts successfully, the information must come
from an institution that enjoys broad approval and public confidence, or perhaps a trusted
individual. Future studies should further explore the circumstances under which institutional
legitimacy effectively mitigates individuals’ tendency to reject facts that are inconsistent with their
prior beliefs.
4. Exploring Alternatives to Legal Discourse
Experiment II demonstrated that legal discourse is ineffective in mobilizing for domestic
condemnation and sanctioning of in-group offenders. It also suggested that a moral discourse may
do a better job at influencing attitudes. The manipulation I used here was very subtle: I only
replaced the word ‘law’ with the word ‘moral’. The conclusion itself was still constructed in a very
‘legalistic’ structure. I did not mention the victims, did not use photos, videos, poems, or other
emotion-triggering manipulations. The finding that moral discourse is more effective than legal
discourse in influencing attitudes on war crimes is powerful, because it demonstrates that even
when keeping the technical and formalistic structure of legal reports – and the exact same standards
– individuals are more likely to be influenced by the report and to express empathy to the out-
group victims when we frame the findings as moral, rather than legal. True, the effect size was
quite small. But the fact that it mattered at all is remarkable, considering the subtlety of the
manipulation.
A possible explanation for this finding is the alienating effect of legal discourse, and the
perception of international law as a technical and marginalizing tool that speaks only to a small
group of experts. In contrast, moral arguments are perceived as intuitive and inclusive, and
encourage wider participation by society members. As Martha Minow suggests, maybe words by
lawyers differ from words by journalists and artists, and maybe “the more intimate and yet more
widely accessible languages of literature and popular music lyrics may change minds and prompt
actions.”160 This study is only the beginning of an exploration of the unique effects of the
legalization of truth on individuals’ beliefs and attitudes. In future studies I will continue to explore
alternatives to legal discourse, and will measure their impact on attitudes and beliefs, including
160 Minow, supra note 130, at 1673, 1685.
Krebs The Legalization of Truth
52
empathy for out-group members. In the following section I suggest several alternatives to existing
structures that stem from these findings, as well as concrete recommendations for future fact-
finding efforts regarding the Kunduz hospital bombing case.
F. RECOMMENDATIONS: REDESIGNING INTERNATIONAL FACT-FINDING
The findings of the two experiments discussed in this article, together with the
interdisciplinary literature presented above, suggest that the decision to center international fact-
finding efforts around legal discourse – including legal judgment, terminology, and frame – may
be counter-productive. My research indicates that the legal judgment of the fact-finding report is
likely to trigger ideological-based belief polarization; that ‘hot’ legal terminology tends to reduce
the perceived fairness of the report; and that the legal frame appears to be less effective than moral
frame in influencing attitudes. Based on these findings, international organizations, including the
U.N., should rethink the use and design of legal fact-finding missions. First, it is important to
acknowledge the limitations of legal fact-finding and to recognize its questionable efficacy in
promoting its main objectives, mainly, to produce and disseminate facts, and to mobilize domestic
support for accountability measures. Second, responding to these challenges requires both
reforming the goals, structures, and processes of international fact-finding, and a systematic
consideration of alternatives to this method, including formal tribunals or collaborative measures.
While this article is mainly focused on the unintended consequences of legal fact-finding and not
on the potential spectrum of alternative institutional choices, the following section offers several
observations from an institutional design perspective, and proposes recommendations to improve
the design and outcomes of fact-finding bodies. More work is needed to further explore the full
range of alternatives to legal fact-finding, to measure the social impact of these alternatives on
attitudes and beliefs, and to compare the relative pros and cons of each method, in comparison to
legal fact-finding.161
161 In another paper I offer a taxonomy of fact-finding goals, structures, and processes, and provide a systematic
method, based on dispute system design and organizations literature, to match goals with the processes and structures
that are most likely to achieve them.
Krebs The Legalization of Truth
53
1. International Fact-Finding Goals and Purposes
In developing a research agenda for studying the effectiveness of international courts,
Yuval Shany proposed a goal-based approach to evaluate international tribunals.162 Drawing on
the work of the influential sociologist Charles Perrow, Shany offered an analytical framework to
organize and prioritize institutional goals, by distinguishing between official and operational goals,
and by analyzing the goals’ source (external or internal), hierarchical level (ultimate ends or
intermediate), and method or articulation (explicit, implicit, or unstated).163 Building on Shany’s
work, this article suggests employing the goal-based analytical framework not only for evaluating
the effectiveness of international institutions (including fact-finding bodies), but also – and more
importantly – for selecting institutional designs. That is to say, when designing an international
fact-finding mission, the mandating organization should define, clarify, and prioritize the goals
and purposes of this mission. Based on the mission’s main goals, alternative processes and
structures should be considered, matching goals to processes in order to maximize the mission’s
efficacy. Instead of adopting a ‘one size fits all’ approach, international fact-finding would benefit
from careful consideration of alternative processes and structures, and from a learned process of
tailoring concrete processes and structures to specific goals.
As mentioned earlier, U.N. fact-finding missions have been established to fulfill an array
of goals and purposes, including creating a historical record, encouraging domestic accountability,
fostering reconciliation, and preventing future abuses. The structure and processes of fact-finding
in most of these cases were similar: the missions adopted legal discourse and made nonbinding
recommendations. The experimental findings presented above suggest that legal discourse may
not be a suitable method to achieve some of these goals, including creating and disseminating a
unified historical record and promoting domestic accountability: the legal judgment and legal
terminology triggers cognitive and emotional biases that prevent the creation of a collective
historical narrative; and the legal frame is less effective than a moral frame for mobilizing domestic
support for accountability measures (including penalizing offenders and reparations for the
victims). These findings suggest that the goals and processes of international fact-finding should
162 Yuval Shany, Assessing the Effectiveness of International Courts: A Goal-Based Approach, 106 AM. J. INT’L L.
225, 270 (2012). 163 Id. at 232.
Krebs The Legalization of Truth
54
be better studied and understood, and support the creation of a systematic framework to design
fact-finding missions in order to increase their positive social impact.
There are several reasons to support a goal-based approach for the design of international
fact-finding. International fact-finding missions do not enjoy the formal documentation and
organizational sophistication inherent to other international institutions. They do not have charters,
statutes, or rules of procedure and evidence to turn to for guidance. Their mandates are often
crafted hastily, while atrocities are ongoing, without identifying and prioritizing concrete goals.164
Without clarifying the goals and purposes of fact-finding efforts, it is difficult to systematically
consider alternative processes and structures, and to choose, from a variety of options, those most
suitable to promote the mission’s goals. Moreover, lack of clarity concerning the mission’s goals
encourages the adoption of existing or familiar structures, without proper consideration of the
appropriateness of these structures to the current situation or the existence of alternative structures.
For example, the decision of the Human Rights Council to dispatch a fact-finding mission
to the Central African Republic [CAR] was short and laconic, stating only that the Council requests
the High Commissioner “to submit to the Human Rights Council, at its twenty-fourth session, an
interim report on the human rights situation in the Central African Republic.”165 In response, the
U.N. High Commissioner for Human Rights deployed, from June 20 to July 11, 2013, a fact-
finding mission to the CAR to collect information on human rights violations committed in Bangui
and other localities between December 2012 and July 2013. The mission interpreted the vague
mandate to ‘submit a report on the human rights situation’ as directed at legal discourse and
accountability, and, accordingly, submitted a legal report analyzing the applicable law and finding
both parties responsible for gross human rights violations and, possibly, war crimes.166 Neither of
these documents – the mandating resolution or the final report – mentioned the goals of this process
or reasons to adopt a legal analysis, given the mission’s lack of any enforcement capabilities. From
the materials available, it seems that this structure was adopted as a default solution, without
164 See, for example, the UNHRC resolution concerning Burundi. UNHRC Res. S-24/1, supra note 35,
http://www.ohchr.org/EN/HRBodies/HRC/UNIIB/Pages/UNIIB.aspx 165 U.N. HUMAN RIGHTS COUNCIL RES. A/HRC/RES/23/18, Technical Assistance to the Central African Republic in
the Field of Human Rights (June 27, 2013), http://www.un.org/ga/search/view_doc.asp?symbol=A/HRC/RES/23/18 166 Human Rights Council, Report of the United Nations High Commissioner for Human Rights on the Situation of
Human Rights in the Central African Republic, A/HRC/24/59 (Sept. 12, 2013),
http://www.un.org/ga/search/view_doc.asp?symbol=A/HRC/24/59.
Krebs The Legalization of Truth
55
discussing the purposes of dispatching this mission and without considering alternative structures
and processes. This goal ambiguity harms fact-finding efforts, as it encourages practices that are
not necessarily productive or fruitful; it impedes attempts to clarify the policy direction ex ante,
and to assess the mechanism's effectiveness ex post.
Another reason to support a goal-based approach to the design of international fact-
finding missions is that even when goals are mentioned in the mandating process they are often
mixed, sometimes conflicting. The documents concerning the United Nations Independent
Investigation on Burundi [UNIIB], for example, mention several goals and purposes for the
investigation, including ‘preventing further deterioration of the human rights situation,’ making
recommendations ‘on the improvement of the human rights situation’ in Burundi, assisting
reconciliation efforts, ensuring ‘accountability for human rights violations and abuses, including
by identifying alleged perpetrators,’ adopting ‘appropriate transitional justice measures,’ and
issuing a final report and participating in an enhanced interactive dialogue on the human rights
situation in Burundi.167 While all of these goals are valuable and important, it seems unlikely that
a fact-finding mission could accomplish them all at the same time, relying on a single structure. A
tension exists, for example, between the desire to promote accountability by identifying and
prosecuting responsible individuals and the desire to prevent future abuses and to promote
reconciliation, which can sometime be achieved only by promising powerful leaders full or partial
amnesty.168 Some transitional justice measures indeed promote truth and reconciliation processes,
in which participants are promised immunity from prosecution in return for their detailed account
of what happened.
In his oral remarks from March 22, 2016, Christof Heyns, U.N. Special Rapporteur on
extrajudicial, summary, or arbitrary executions and member of the UNIIB, mentioned another
goal, stating that “it is crucial to ascertain the disputed facts in an indisputable manner.”169 As the
167 See UNHRC Res. S-24/1, supra note 35. 168 Andrea Kupfer Schneider, The Intersection of Dispute Systems Design and Transitional Justice, 14 HARV.
NEGOT. L. REV. 289, 292 (2009); see also Jane E. Stromseth et al., CAN MIGHT MAKE RIGHTS?: BUILDING THE RULE
OF LAW AFTER MILITARY INTERVENTIONS 253 (2006). 169 Oral update by Christof Heyns, U.N. Special Rapporteur on extrajudicial, summary or arbitrary executions,
Member of the United Nations Independent Investigation on Burundi (UNIIB), on the Enhanced Interactive
Dialogue on Burundi, HUMAN RIGHTS COUNCIL (Mar. 22, 2016),:
http://libraryresources.unog.ch/c.php?g=462695&p=3162785#21493049
Krebs The Legalization of Truth
56
theoretical framework and the experimental findings described above demonstrate, determining
facts in an indisputable manner is probably not a productive way to think about the process of
truth-telling and truth-seeking concerning intense social controversies. Moreover, the desire to
present facts as indisputable may conflict with conflict resolution and reconciliation efforts, which
are often based on a narrative or restorative approaches to truth – rather than on binary legal
discourse – embracing the coexistence of conflicting narratives and different descriptions or
experiences of the same events. Additionally, thinking of determining facts in an indisputable way
raises another challenge, as it relates to two separate goals: determining the facts, and
disseminating these facts successfully to a variety of heterogeneous sub-groups. To achieve this
dual goal, documenting abuses is not enough; it is just as important to contribute to the creation of
a shared narrative. As this article demonstrates, the processes and structures adopted to find the
truth inevitably influence the persuasive power of these facts and determine the success or failure
of their dissemination. Since different aspects of the legal discourse (including judgment and
terminology) trigger cognitive and emotional biases and selective information processing,
successful dissemination of facts could potentially benefit from a different approach to fact-finding
and truth. The following section will suggest some alternative approaches.
2. Alternatives to Legal Discourse
The popularity of legal discourse in the context of fact-finding should not be mistaken as a
force of nature. Instead, it should be understood as what it is: a choice; a social structure that
constructs a specific interpretation of reality. As Jack Balkin points out, “law’s truth is not the only
truth, and law’s vision of reality is not the only reality. Law’s power to enforce its vision of the
world can clash with other practices of knowledge, and with other forms of truth.”170 There are
many other forms of knowledge acquisition in the world, and many other conventions through
which people try to determine what is true or false, try to gather and assess knowledge.171 In his
canonic book, What is History, Edward Carr explains that history is a “continuous process of
interaction between the historian and his facts, an unending dialogue between the present and the
past.”172 Unlike history-making, the traditional criminal legal process is final and binary. The
170 Jack M. Balkin, The Proliferation of Legal Truth, 26 HARV. J. L. & PUB. POL’Y 5, 7 (2003). 171 Id. at 11. 172 EDWARD H. CARR, WHAT IS HISTORY? 35 (1961).
Krebs The Legalization of Truth
57
judgment can be either incriminating or absolving, not both. Conflicting narratives cannot co-exist.
This unavoidable affirmation of one narrative and rejection of others triggers, as explained earlier,
cognitive bias and rejection or distortion of information.
To overcome some of these unintended consequences of legal discourse and to provide a
more inclusive and nuanced account of what happened, other types of discourse can be adopted,
promoting different interpretations of ‘truth.’ While exploring the broad range of potential
discourses and the types of realities or ‘truths’ that they produce is beyond the scope of this article,
I will mention just a few alternatives, adopted by other transitional justice mechanisms in the past.
The South African Truth and Reconciliation Commission, for example, included four different
types of ‘truth’ in its fact-finding processes: a forensic truth, focused on the objective, physical,
information; a narrative truth, focused on the personal stories and experiences of individuals –
both victims and perpetrators – and on creation of united, restored memories; a social truth,
established through interaction, discussion and debate; and a healing/restorative truth, enabled
through public acknowledgment and common memories of the events.173 The medical and
psychological literature has also revealed the potential contribution of this broader understanding
of truth for therapy;174 and socio-psychological research has demonstrated that third parties can
potentially play a valuable role in the process of reevaluating social narratives and beliefs.175
While courts of law must rely on legal discourse and cannot escape making final and binary
determinations based on legal judgment, terminology, and frame, international fact-finding efforts
may, in some circumstances, benefit from considering other types of discourse, depending on the
primary goals of the mission.176 A broader, non-legalistic interpretation of truth, such as a
173 Alex Boraine, Truth and Reconciliation in South Africa: The Third Way, in TRUTH V. JUSTICE: THE MORALITY
OF TRUTH COMMISSIONS 141-157, 151-153 (Robert I. Rotberg & Dennis Thompson eds., 2000). 174 Alan W. Schefin, Narrative Truth, Historical Truth, and Forensic Truth, in THE MENTAL HEALTH PRACTITIONER
AND THE LAW: A COMPREHENSIVE HANDBOOK (Lawrence E. Lifson & Robert I. Simon eds., 1998). 175 Bar-Tal, supra note 27, at 22. Others found that third-party activities to open or maintain lines of communication
are the most consistently effective conflict management techniques for preventing escalation. William J. Dixon,
Third-Party Techniques for Preventing Conflict Escalation and Promoting Peaceful Settlement, 50 INT’L ORG. 653,
671 (1996). 176 Marco Sassoli, for example, advised the international community to avoid using international criminal law as an
alibi for failing to engage in cooperative and diplomatic efforts to resolve the conflict itself. Marco Sassoli, The
Implementation of International Humanitarian Law: Current and Inherent Challenges, 10 Y.B. INT’L
HUMANITARIAN L. 45, 55 (2007).
Krebs The Legalization of Truth
58
narrative, social, or restorative truth,177 may be better suited to contribute to a collective narrative
of contested events, create a ‘shared history,’ and disseminate otherwise threatening facts.
Additionally, legal discourse, especially in the context of criminal law and accountability,
is focused on individualized blame.178 While individualizing guilt serves several purposes, it has
its own problems and dangers. As Barbara Fried points out, “we have gotten nothing from our
40-year blame fest except the guilty pleasure of reproaching others for acts that, but for the
grace of God, or luck, or social or biological forces, we might well have committed
ourselves.”179 Discussing the South African Truth and Reconciliation Commission, Gibson
argued that promoting an alternative of “shared blame” was the single most important
characteristic of the South African truth and reconciliation process.180 As he explains, “sharing
responsibility, blame, and victimhood creates a common identity, which can provide a basis for
dialogue. If people are no longer dogmatically attached to a “good versus evil” view of the struggle,
then perhaps a space for reconciliation is opened.”181 Replacing the criminal legal discourse of
individualized guilt with a social discourse of “shared blame,” could potentially alleviate the
binary legal discourse of guilt/innocence into a constructive social discourse supporting
reconciliation.
Alternatively, the scientific and medical literature has challenged the criminalization of
human errors and the culture of blame altogether.182 In his book, Just Culture: Balancing Safety
and Accountability, Sidney Dekker concludes that the threat of judicial involvement is enough to
prevent people from coming forward with information about an incident that they were involved
in.183 Judicial involvement, he argues, can engender a climate of fear and silence, in which it can
177 Boraine, supra note 173, at 151-153. 178 MICHAEL S. MOORE, PLACING BLAME: A THEORY OF THE CRIMINAL LAW (2010); Payam Akhavan, Beyond
Impunity: Can International Criminal Justice Prevent Future Atrocities?, 95 AM. J. INT’L L. 7 (2001). 179 Barbara H. Fried, Beyond Blame: Would We Be Better Off in a World Without Blame?, BOSTON REV., June 28,
2013. 180 James L. Gibson, The Contributions of Truth to Reconciliation Lessons from South Africa, 50 J. CONFLICT
RESOL. 409, 417 (2006). 181 Id. 182 Geoffrey Thomas, A crime against safety, 44(1) AIR TRANSPORT WORLD 57 (2007) (calling to change the
healthcare culture of blame, as it only hinders investigations into why an error occurred and how to prevent future
ones); Lucian L. Leape et al, Promoting Patient Safety by Preventing Medical Error, 280(16) JAMA 1444 (1998)
(arguing that escalating punishments for errors has been less than successful, resulting in suppression, stonewalling,
and cover-up by clinicians and healthcare organizations). 183 SIDNEY DEKKER, JUST CULTURE: BALANCING SAFETY AND ACCOUNTABILITY 103-105 (2012).
Krebs The Legalization of Truth
59
be difficult, if not impossible, to get access to information that may be critical to finding out what
happened and to prevent similar errors in the future.184 Others found, similarly, that cultures of
blame – intensified by lawyers and the media – can lead individuals and organizations to blame
others rather than take responsibility for the error and thereby explore and create solutions to
address the error or problem.185
Therefore, official policy-making bodies and experts in medical and human error have
called for a shift in perspective to a blame-free culture within medicine, predicated on the basis
that errors are largely attributable to systems rather than individuals.186 While, arguably, healthcare
and aviation errors are different than military ones, at least to some extent, an important similarity
relates to the psychological processes leading individuals to share information, to take
responsibility, and to find solutions and prevent repetition of the same errors time and again.187 If
criminalization and blame have indeed a detrimental effect on willingness to report and disclose
information, then we should be motivated to explore blame-free alternatives to fact-finding, which
could, potentially, motivate individuals to share information and experiences openly.
3. The Kunduz Hospital Bombing: Finding the Facts, Preventing Future Attacks
Let us now return to the attack on the MSF hospital in Kunduz. At the time of this writing,
both MSF and UNAMA, as well as several other human rights organizations, continue to demand
the establishment of a legalized fact-finding mission to investigate the attack (or to task the existing
IHFFC with this responsibility).188 It seems that the international community has been trapped in
184 SIDNEY DEKKER, JUST CULTURE: BALANCING SAFETY AND ACCOUNTABILITY 103-105 (2012). And see, also,
Sidney Dekker, The criminalization of human error in aviation and healthcare: A review, 49(2) SAFETY SCIENCE
121 (2011) (exploring the social causes and psychological and organizational consequences of the criminalization of
human error in aviation and healthcare and concluding that criminal prosecution may threaten safety, as it has a
detrimental effect on willingness to report and disclose safety-related information). 185 Jamie Dickey, Ralph J. Damiano, and Ross Ungerleider, Our Surgical Culture of Blame: a Time for Change,
126(5) THE J. OF THORACIC AND CARDIOVASCULAR SUR. 1259 (2003). 186 Molly E. Collins et al, On the Prospects for a Blame-Free Medical Culture, 69(9) SOCIAL SCIENCE & MEDICINE
1287 (2009). See, also, Richard T. Penson et al, Medical Mistakes: a Workshop on Personal Perspectives, 6(1) THE
ONCOLOGIST 92 (2001) (finding that openly sharing experiences in a confidential setting helps defuse feelings of
guilt and challenges the culture of shame and isolation that often surrounds medical errors). 187 For a brief typology of human errors and failures, see: Amy C. Edmondson, Strategies for learning from failure,
89(4) HARVARD BUSINESS REVIEW 48 (2011). 188 See, for example, the online petition issued by MSF: Tell U.S. President Obama to Consent to Independent
Investigation of Kunduz Hospital Bombing, which recruited by July 28, 2016 over 559,000 signatures,
https://www.change.org/p/barack-obama-kunduz-investigation. The petition calls for “an independent international
Krebs The Legalization of Truth
60
the misconception that legal fact-finding is the only method to resolve factual controversies and
mobilize for action. However, as this article demonstrates, legal discourse is ineffective in realizing
both. Applying the previous discussion to the case of the Kunduz hospital airstrike, I make the
following suggestions and observations:
First, prioritizing goals and purposes: what is more important – finding what happened, or
determining who is responsible? Prosecuting those involved, or implementing long-term
institutional changes? Of course, it is possible to envisage a combination of goals to be pursued in
a single case. However, clarifying and prioritizing these goals would be helpful in deciding which
structures and processes would best fit the particular situation. Based on MSF’s communications,
reports, and demands, it seems that the organization’s main goal is not individualized prosecutions
and adjudication, but rather finding out what really happened – what caused the U.S. military to
mistakenly identify a hospital as a military target, and accordingly to attack it. They seek a way to
determine and disseminate facts successfully, reconciling conflicting accounts concerning what
happened.189
Second, identifying alternatives to legal discourse: if, indeed, the main goals of the fact-
finding efforts are determining and disseminating facts and producing an agreed-upon record of
events, as well as preventing future breaches of the protections accorded to medical objects, the
adoption of legal discourse might be counter-productive. The binary legal judgment – assigning
blame and incriminating or absolving specific individuals – might stand in the way of encouraging
information sharing and creating a shared understanding of what happened. Legal discourse creates
an adversarial approach to truth, in which conflicting narratives clash and only one prevails. If, at
the end of an international fact-finding process, the MSF narrative prevails, it is conceivable that
the U.S. military would continue to reject this narrative, especially considering the nonbinding
nature and lack of enforcement capacities of fact-finding reports. If the narrative of the U.S.
investigation into the events of October 3 by the International Humanitarian Fact-Finding Commission (IHFFC)”,
which will reaffirm the commitment to international humanitarian law and the right to provide medical care in
conflict zones. See, also, Afghanistan: US Should Conduct Criminal Inquiry in MSF Attack, HUMAN RIGHTS WATCH
(Dec. 21, 2015), https://www.hrw.org/news/2015/12/21/afghanistan-us-should-conduct-criminal-inquiry-msf-attack. 189 See, e.g., Doctors Without Borders, supra note 9 Doctors Without Borders, Kunduz: Some of MSF’s Questions in
Response to the U.S. Military Investigation into Their Attack on the Hospital (Apr. 29, 2016),
http://www.msf.org/en/article/kunduz-some-msf%E2%80%99s-questions-response-us-military-investigation-their-
attack-hospital.
Krebs The Legalization of Truth
61
military prevails, concluding that the named individuals did not commit war crimes, these
individualized findings might mask broader systemic failures.
To find the truth and reconcile conflicting narratives, MSF should consider forsaking its
commitment to legal judgment and blame, which has detrimental effects on willingness to report
and disclose sensitive information concerning erroneous risk assessments and organizational
failures. Alternatively, an organizational discourse, promoting a “learning from failure” approach,
should be considered, offering collaborative, blameless, fact-finding structures to motivate
information sharing and disclosure, as well as organizational reform.190 Additionally, an
organizational discourse would shift the focus of attention from issues relating to individual
criminal responsibility to broader social processes, including organizational culture, decision-
making processes, and structural biases leading to erroneous risk assessments, which may better
account for some of the failures leading to the attack on the MSF hospital. While analyzing the
effects of various alternatives to legal discourse, including an organizational discourse, is outside
the scope of this article, the literature surveyed here testifies to the potential of such alternative
discourses in instigating better practices of information sharing and organizational change.
G. CAVEATS AND FUTURE RESEARCH
Experimental methods have unique value for studying the efficacy of legal institutions. By
randomly assigning treatments, the researcher is able to control the data-generation process and to
estimate the influence of a specific intervention on an outcome.191 Nonetheless, this method is not
without flaws.
First, this article demonstrates that individuals respond in different ways to new
information, depending on their national, social, cultural, and political affiliations. Therefore, the
findings of the experiments described in this article are limited to the social groups that were
included in the experiments, mainly American Nationals. It is possible that different social groups
hold different beliefs and respond differently to international law and international institutions. To
support a more inclusive and generalizable argument concerning the impact of international law
190 Amy C. Edmondson, Strategies for Learning from Failure, 89 HARV. BUS. REV.48 (2011). 191 Chilton & Tingley, supra note 19, at 6.
Krebs The Legalization of Truth
62
on domestic attitudes, these hypotheses must be tested using representative samples of other
nations. In future articles I will report the results of similar experiments fielded in Israel and
Canada. These comparative experiments will enable a comparison of the impact of international
fact-finding reports on public beliefs and attitudes in different societies.
Second, the findings of the experiments described in this paper are also limited to the
specific situation which was chosen for the vignette: a small-scale war crime, which could lead to
different legal and moral conclusions. While this scenario was carefully chosen to represent a real-
life, believable situation, allowing subjects to reach their own judgments and to reduce social
desirability bias, other situations could result in different outcomes. Future research should test the
impact of international law on domestic attitudes when in-group offenders are blamed for genocide
or other large-scale grave violations of international law.
Third, Experiment I specifically measured the impact of ‘war crimes’ terminology on the
perceived credibility of fact-finding reports. This wording was chosen to represent realistic use of
this legal term, as it frequently appears in international fact-finding reports. The term represents
‘hot’ legal terminology with negative moral value. While the experiment successfully
demonstrates the negative impact this specific term has on the perceived fairness of fact-finding
reports, the results may also be motivated by remedial deterrence (or solution aversion). That is to
say, the severe legal and moral implications stemming from war crimes accusations may motivate
subjects to discredit the information altogether or refuse to sanction the offenders. Additionally,
the discrepancy between the legal and political meanings of the term ‘war crimes’ could have
influenced some participants to respond negatively to the treatment simply because they thought
the described events are not grave enough to be considered ‘war crimes.’ To test the exact
mechanism or reason for the negative effect found in this experiment – whether it resulted from
an identity-based bias, solution aversion, or discrepancy between the legal and political meaning
of war crimes – more research should be conducted, varying the gravity of the crimes, as well as
the anticipated sanction.
Fourth, individuals typically receive this type of information from the media. A more
realistic scenario would have presented the information to the research subjects as a news article.
However, research shows that the specific media outlet conveying the information influences
perceived credibility and assessments of new information. Even without mentioning the specific
Krebs The Legalization of Truth
63
outlet, subjects’ authentic reactions to the report might be contaminated with their views of the
‘media.’ Therefore, I decided to use a direct ‘quote’ of the executive summary of the report rather
than a news article reporting its main finding. This design choice allowed me to isolate subjects’
reactions to the report rather than mix it with media bias.
Finally, public opinion or domestic attitudes are not necessarily a proxy for policy or
behavioral changes. This is particularly true in non-democratic states, but is also relevant for
democracies. Having said that, this article specifically tests domestic attitudes because I believe
these are particularly important for international fact-finding efforts. Increasingly, with the
development of social media and alternative information channels, social sub-groups hold
conflicting beliefs concerning physical events. How many people were killed? What was the cause
of death? Were the victims militants or civilians? And based on different assessments of what
actually happened, individuals form opinions on specific responses to the relevant events, but also
on the conflict, its roots, the enemy, and possible solutions. Regardless of potential policy or
behavioral outcomes, I argue that domestic attitudes on war crimes matter because they serve as
societal foundations which preserve and even intensify conflicts. And it is our duty to design
institutions that can potentially mitigate, rather than intensify, this problem.
H. CONCLUSION
War crimes investigations are intended, among other things, to assert and disseminate the
truth and to prevent future atrocities. Unfortunately, modern international fact-finding missions
adopt legal discourse to construct and interpret facts, and this discourse tends to trigger cognitive
and emotional biases which may frustrate attempts to achieve these exact goals. An adversarial,
binary, legal-judgment-approach to truth might frustrate attempts at resolving factual controversies
and disseminating threatening information. ‘Hot’ legal terminology, such as ‘war crimes,’ might
trigger defensive reaction of rejection of information and decrease the perceived fairness of the
report. A legal frame fails to influence domestic attitudes or to mobilize support for sanctioning
and reparations. The choice to ascertain and disseminate truth using legal discourse is not without
benefits – it provides legal rules, terminology, and conventions to structure and guide the
production of facts and construction of meaning. However, at the same time, it triggers cognitive
Krebs The Legalization of Truth
64
and emotional biases that frustrate efforts to disseminate controversial information and to resolve
factual disputes; and it lacks the emotional appeal, participatory value, and social cues that moral
expressions or other types of social truth-telling entail.
The data reported here suggest that sometimes more information means less shared
knowledge: while opposing social and political groups could potentially agree on factual findings
such as numbers of victims or the timeframe of a military attack, the adoption of binary legal
judgment and politicized legal terminology contributes to a bitter divide in public opinion
concerning war crimes. Ironically, the tendency to turn to legal discourse to achieve factual
accuracy and accountability triggers social dynamics that frustrate both objectives. In order to have
a stronger, more robust, persuasive value, fact-finding missions should rethink their focus on legal
discourse and consider replacing it with other, more socially oriented forms of truth-telling and
communication. By widening the corpus of relevant facts, by restructuring the boundaries of what
is true or false, and by accepting the potential co-existence of conflicting narratives, fact-finding
mechanisms may be able to dismantle some of the biases and distortion, and to accomplish their
goals.
Krebs The Legalization of Truth
65
I. APPENDICES
1. U.N. Fact-Finding Missions – Goal Dataset
Goals Legend/Coding (1=goal mentioned in the mission’s mandate, 0=goal not mentioned):
IL = Investigating International Law Violations
Fact = Establishing the Facts
Truth = Establishing the Truth
Prev = Preventing Future Atrocities
Rec = Promoting Reconciliation
Account = Promoting Accountability
Assist = Assisting Domestic Authorities
Year Investigated
Country
Mandating
Authority
Goals
IL Fact Truth Prev Rec Account Assist
2015 Burundi HRC 1 0 0 1 1 1 1
2015 Syria HRC 1 0 0 0 0 1 0
2015 Libya HRC 1 1 0 0 0 1 0
2015 South Sudan HRC 1 0 0 0 0 1 0
2015 Boko Haram HRC 1 0 0 0 0 0 0
2014 Syria HRC 1 0 0 0 0 1 0
2014 Iraq HRC 1 1 0 0 0 1 0
2014 Sri-Lanka HRC 1 1 0 0 0 1 0
2014 Israel/Palestine HRC 1 1 0 1 0 1 0
2014 Eritrea HRC 1 0 0 0 0 1 0
2013 DPRK HRC 1 0 0 0 0 1 0
2013 Syria HRC 1 0 0 0 0 1 0
2013 Syria HRC 0 1 0 0 0 0 0
2013 CAR UNSC 1 0 0 0 0 1 0
2013 Mali HRC 1 0 0 0 0 0 0
2013 CAR HRC 1 0 0 0 0 0 0
2012 Syria HRC 1 0 0 0 0 1 0
2012 Syria HRC 1 0 0 0 0 1 0
2011 Syria HRC 1 1 0 0 0 1 0
2011 Syria HRC 1 1 0 0 0 1 0
2012 Israel/Palestine HRC 1 1 0 0 0 0 0
2011 Côte d’Ivoire HRC 1 0 0 0 0 1 0
2011 Libya HRC 1 0 0 0 0 1 0
2010 Israel/Turkey HRC 1 0 0 0 0 1 0
2010 Sri-Lanka SG 1 0 0 0 0 1 0
2010 Israel/Palestine HRC 1 1 0 0 0 1 0
2009 Israel/Palestine HRC 1 0 0 0 0 0 0
Krebs The Legalization of Truth
66
Year Investigated
Country
Mandating
Authority
Goals
IL Fact Truth Prev Rec Account Assist
2009 Honduras HRC 1 0 0 0 0 0 0
2009 Nepal HCHR 1 0 0 0 0 0 0
2009 Guinea SG 1 1 0 0 0 1 0
2008 Congo SG 1 0 0 0 0 0 1
2008 Kenya HCHR 1 0 0 0 0 0 0
2006 Israel/Palestine HRC 0 0 0 1 0 1 0
2006 Timor SG 1 1 0 0 0 1 0
2006 W. Sahara HCHR 1 0 0 1 0 0 0
2006 Israel/Lebanon HRC 1 1 0 0 0 0 0
2006 Afghanistan HCHR 1 0 0 0 0 0 0
2006 Darfur HRC 1 0 0 0 0 0 0
2005 Uzbekistan HCHR 1 1 0 0 0 0 0
2005 Togo HCHR 1 0 0 0 0 0 0
2004 Darfur HCHR 1 0 0 0 0 0 0
2004 Côte d’Ivoire UNSC 1 0 0 0 0 1 0
2004 Abidjan SG 1 0 0 0 0 1 0
2004 Darfur UNSC 1 0 0 0 0 1 0
2002 Columbia HCHR 1 0 0 0 0 0 0
2002 Côte d'Ivoire SG 1 0 0 0 0 0 0
2000 Togo GA 1 0 1 0 0 0 0
2000 Côte d'Ivoire SG 1 0 0 0 0 0 0
2000 Israel/Palestine CHR 1 0 0 1 0 0 0
1999 Afghanistan GA 1 0 0 0 0 0 0
1999 Timor CHR 1 1 0 0 0 0 0
1999 Cambodia GA 1 0 0 0 1 1 1
1998 Afghanistan HCHR 1 0 0 0 0 0 0
1997 Congo CHR 1 0 0 0 0 1 0
1997 Congo SG 1 0 0 0 0 1 0
1995 Burundi UNSC 1 1 0 0 1 1 0
1994 Rwanda UNSC 1 0 0 0 0 0 0
1993 Georgia SG 1 0 0 0 0 0 0
1993 Burundi SG 1 1 0 0 0 0 0
1993 Liberia UNSC 1 0 0 1 0 1 0
1992 Yugoslavia UNSC 1 0 0 0 0 0 0
1988 Cuba CHR 1 0 0 0 0 0 0
1979 Israel/Palestine UNSC 1 0 0 0 0 1 0
1975 Chile CHR 1 0 0 0 0 0 0
1973 Mozambique GA 1 0 0 0 0 0 0
1963 Vietnam GA 0 1 0 0 0 0 0
Total: 66 63 17 1 6 3 33 3
Krebs The Legalization of Truth
67
2. Experiments’ Sampling and Demographics
(a) Matching and Weighting for experiment I:
YouGov interviewed 2251 respondents who were then matched down to a sample of 2000 to
produce the final dataset. The respondents were matched to a sampling frame on gender, age, race,
education, party identification, ideology, and political interest. The frame was constructed by
stratified sampling from the full 2010 American Community Survey (ACS) sample with selection
within strata by weighted sampling with replacements (using the person weights on the public use
file). Data on voter registration status and turnout were matched to this frame using the November
2010 Current Population Survey.
Data on interest in politics and party identification were then matched to this frame from the
2007 Pew Religious Life Survey. The matched cases were weighted to the sampling frame using
propensity scores. The matched cases and the frame were combined and a logistic regression was
estimated for inclusion in the frame. The propensity score function included age, gender,
race/ethnicity, years of education, and ideology. The propensity scores were grouped into deciles
of the estimated propensity score in the frame and post-stratified according to these deciles.
Table 1. Sample Descriptive Statistics (experiment I)
Proportion/mean
Experimental treatments
Judgment
Incriminating legal judgment 0.322
War crimes judgment 0.341
Absolving legal judgment 0.338
Institution
United Nations 0.335
U.S. Military Court 0.332
U.S. Federal Court 0.334
Covariates
Ideology
Conservative 0.366
Moderate 0.325
Liberal 0.256
Gender (female) 0.526
Age 49 (s.d.=16.369,
min=19, max=93)
Race
While 0.717
Krebs The Legalization of Truth
68
Black 0.108
Hispanic 0.123
Other 0.055
Education
No high school education 0.037
High school education 0.345
College education (or partial) 0.511
Graduate education 0.108
Valid voter registration status 0.876
(b) Matching and Weighting for experiment II:
YouGov interviewed 1152 respondents who were then matched down to a sample of 1000 to
produce the final dataset. The respondents were matched on gender, age, race, education, party
identification, ideology, and political interest.
YouGov then weighted the matched set of survey respondents to known characteristics of the
general population of the United States from the 2010 American Community Survey and 2012
Current Population Survey.
Table 2. Sample Descriptive Statistics (experiment II)
Proportion/mean
Experimental treatments
Judgment/framing
Incriminating legal judgment 0.199
Absolving legal judgment 0.206
Incriminating moral judgment 0.213
Absolving moral judgment 0.197
Control 0.185
Covariates
Ideology
Conservative 0.348
Moderate 0.3
Liberal 0.265
Gender (female) 0.52
Age 47 (s.d.=16.998,
min=18, max=93)
Race
While 0.721
Black 0.111
Hispanic 0.111
Other 0.057
Krebs The Legalization of Truth
69
Education
No high school education 0.055
High school education 0.377
College education (or partial) 0.482
Graduate education 0.085
Valid voter registration status 0.865
Krebs The Legalization of Truth
70
3. Vignettes and Questionnaires
(a) Experiment I:
Each responded was randomly assigned into one of 9 treatment groups, based on 2
randomized variables:
1. Institution_rand: the United Nations, a U.S. Military Court or a U.S. Federal Court
2. Judgment_rand: The U.S. Marines did not violate international law, the U.S. Marines
violated international law or the U.S. Marines committed war crimes.
First, respondents read the following paragraph:
The following questions are about a military operation in Afghanistan that happened a few
months ago. Initially, it was unclear how many people died, and whether they were Al-Qaeda
members or innocent civilians.
Therefore, the institution_rand appointed a team of experts to investigate.
The summary of their report includes 2 parts (which are reprinted on the next 2 screens): in the
first part the investigators give their account of what happened (the facts of the case). In the
second part the investigators give their legal judgment of these facts, concluding that the United
States judgment_rand.
Please read both parts carefully. Afterwards, we will ask for your opinions about each part of the
report.
Second, the respondents read the two parts of the fact-finding report:
[Insert UN1/MC1/CC1 here]
New Screen
[Insert UN2/UN3/UN4/MC2/MC3/MC4/CC2/CC3/CC4 here]
Third, the respondents were asked to answer the following questions:
“The investigation was conducted because the facts of this case were highly contested
(especially how many people were killed, and whether they were Al-Qaeda members or innocent
civilians).
Krebs The Legalization of Truth
71
Therefore, we want to know your opinion on the credibility of the facts determined by the
investigators (the description of what happened, including the finding that 4 unarmed Afghan
civilians were killed by the Marines):
Do you think that the facts summarized by the investigators are –
1 Completely inaccurate
2 Mostly inaccurate
3 Somewhat inaccurate
4 Somewhat accurate
5 Mostly accurate
6 Completely accurate
In your opinion, were the facts presented by the investigators –
1 Completely biased
2 Mostly biased
3 Somewhat biased
4 Somewhat objective
5 Mostly objective
6 Completely objective
New Screen
Now we want your opinion on the investigators' legal judgment of the events (determining that
the United States did not violate international law/violated international law/committed war
crimes).
In your opinion, was the investigators’ judgment –
1 Completely biased
2 Mostly biased
3 Somewhat biased
4 Somewhat objective
5 Mostly objective
6 Completely objective
New screen
Krebs The Legalization of Truth
72
Based on the summary of the report, we now want to know your opinion on the full report.
In your opinion, was the investigators' report –
1 Completely unfair
2 Mostly unfair
3 Somewhat unfair
4 Somewhat fair
5 Mostly fair
6 Completely fair
New screen
Now we want to know your personal opinion on the events.
How much do you agree or disagree with each of the following statements?
“The US Marines did the best they could under the circumstances.”
1 Disagree strongly
2 Disagree somewhat
3 Disagree slightly
4 Agree slightly
5 Agree somewhat
6 Agree strongly
New screen
“The US Marines should be prosecuted.”
1 Disagree strongly
2 Disagree somewhat
3 Disagree slightly
4 Agree slightly
5 Agree somewhat
6 Agree strongly
Fourth, after completing the survey, the respondents were given various tasks related to other
issues (such as education reforms). After about 10 minutes of unrelated assignments,
respondents were asked the following two attitudinal questions:
Krebs The Legalization of Truth
73
What is your opinion of the institution_rand?
1 Very favorable
2 Somewhat favorable
3 Somewhat unfavorable
4 Very unfavorable
5 Don’t know / Never heard of it
New Screen
How much do you agree or disagree with the following statement?
"Patriots should support America even if it is in the wrong"
1 Disagree strongly
2 Disagree somewhat
3 Agree somewhat
4 Agree strongly
Krebs The Legalization of Truth
74
UN1 (similar to MC1 and CC1, the only difference being the institution’s name and
symbol):
Krebs The Legalization of Truth
75
UN2 (similar to MC2 and CC2, the only difference being the institution’s name and
symbol):
Krebs The Legalization of Truth
76
UN3 (similar to MC3 and CC3, the only difference being the institution’s name and
symbol):
Krebs The Legalization of Truth
77
UN4 (similar to MC4 and CC4, the only difference being the institution’s name and
symbol):
Krebs The Legalization of Truth
78
(b) Experiment II:
Each responded was randomly assigned into one of 4 treatment groups or a control group,
based on 2 randomized variables:
1. Framing_rand: international law standards or universal moral standards.
2. Judgment_rand: The U.S. Marines did not violate or violated the framing_rand.
3. Control group subjects read the same facts as everybody else, but did not receive any
judgment (condemning or absolving, legal or moral).
First, respondents read the following paragraph:
The following questions are about a military operation in Afghanistan which happened a few
months ago. Initially, it was unclear how many people died, and whether they were Al-Qaeda
members or innocent civilians.
Therefore, a team of international experts was asked to investigate.
The summary of their report is reprinted on the next screen. In their summary, the investigators
give their account of what happened (the facts of the case).
The summary of their report includes 2 parts (which are reprinted on the next 2 screens): in the
first part the investigators give their account of what happened (the facts of the case). In the
second part the investigators give their legal judgment of these facts, concluding that
judgment_rand [control group subjects read a similar paragraph that only mentioned the first
part of the report].
Please read the report's summary carefully. Afterwards, we will ask for your opinion about it.
Krebs The Legalization of Truth
79
Nangar Khel (Afghanistan) Report:
Summary of the Facts
On August 4, 2013, a member of Al-Qaeda exploded a bomb near a US Marine base in the
Nangar Kahel region in Afghanistan, killing a US Marine.
Three Marines, who witnessed the explosion, chased the attacker into a nearby marketplace.
They shot their guns repeatedly until they killed the attacker.
Four Afghans who were near the shooting were also killed from the gunshots, and nine others
were wounded. The Marines said they saw weapons and thought these people were Al-Qaeda
members. After examining the evidence, the investigators concluded that these people were
unarmed civilians.
New screen
Second, respondents in the treatment groups read one of the following 4 treatments [2X2
design]:
Nangar Khel (Afghanistan) Report (continued):
Summary of the Legal Conclusions
According to international law (including the Geneva Conventions), soldiers are obligated to
exercise reasonable care to protect civilians during armed conflict.
(1) Incriminating judgment, legal framing:
Having investigated the facts of the case, the investigators concluded that the US Marines
violated this legal standard and that the United States is, therefore, legally responsible for the
death and injury of these civilians.
(2) Absolving judgment, legal framing:
Having investigated the facts of the case, the investigators concluded that the US Marines did not
violate this legal standard and that the United States is, therefore, not legally responsible for
the death and injury of these civilians.
(3) Incriminating judgment, moral framing:
Having investigated the facts of the case, the investigators concluded that the US Marines
violated this moral standard and that the United States is, therefore, morally responsible for
the death and injury of these civilians.
(4) Absolving judgment, moral framing:
Having investigated the facts of the case, the investigators concluded that the US Marines did not
violate this moral standard and that the United States is, therefore, not morally responsible for
the death and injury of these civilians.
Krebs The Legalization of Truth
80
Third, the respondents were asked to answer the following questions:
The investigation was conducted because the facts of this case were highly contested (especially
how many people were killed, and whether they were Al-Qaeda members or innocent civilians).
Therefore, we want to know your opinion on the credibility of the facts determined by the
investigators (the description of what happened):
Do you think that the facts summarized by the investigators are –
1 Completely inaccurate
2 Mostly inaccurate
3 Somewhat inaccurate
4 Somewhat accurate
5 Mostly accurate
6 Completely accurate
In your opinion, were the facts presented by the investigators –
1 Completely biased
2 Mostly biased
3 Somewhat biased
4 Somewhat objective
5 Mostly objective
6 Completely objective
New screen
Now we want your opinion on the investigators' legal/moral judgment of the events
(determining that judgment_rand).
In your opinion, was the investigators’ judgment –
1 Completely biased
2 Mostly biased
3 Somewhat biased
4 Somewhat objective
5 Mostly objective
6 Completely objective
Krebs The Legalization of Truth
81
New screen
Based on the summary of the report, we now want to know your opinion on the full report.
In your opinion, was the investigators' report –
1 Completely unfair
2 Mostly unfair
3 Somewhat unfair
4 Somewhat fair
5 Mostly fair
6 Completely fair
New screen
Now we want to know your personal opinion on the events.
Please tell us how much you agree or disagree with each of the following statements:
“The US Marines did the best they could under the circumstances.”
1 Disagree strongly
2 Disagree somewhat
3 Disagree slightly
4 Agree slightly
5 Agree somewhat
6 Agree strongly
“The US Marines should be prosecuted.”
1 Disagree strongly
2 Disagree somewhat
3 Disagree slightly
4 Agree slightly
5 Agree somewhat
6 Agree strongly
Krebs The Legalization of Truth
82
“The US Government should compensate the Afghan victims.”
1 Disagree strongly
2 Disagree somewhat
3 Disagree slightly
4 Agree slightly
5 Agree somewhat
6 Agree strongly
Krebs The Legalization of Truth
83
4. Summary of Treatment Effects
(a) Experiment I:
Table 3. Summary of Treatment Effects: Mean Report Fairness, by Treatment Groups and Ideology
Judgment Institution All
Respondents
Ideology
Liberals Moderates Conservatives
Absolving United Nations 61.51
(57.93, 65.09)
45.09
(37.31, 52.87)
63.07
(57.98, 68.16)
70.25
(63.97, 76.53)
Military Court 60.5
(57.18, 63.82)
52.3
(44.42, 60.19)
60.41
(55.78, 65.04)
66.75
(61, 72.51)
Federal Court 58.02
(54.59, 61.46)
46.66
(40.39, 52.94)
59.67
(53.97, 65.38)
63.66
(57.44, 69.87)
Total (AJ)
60.05
(58.07, 62.04)
48.02
(43.86, 52.18)
61.1
(58.2, 64)
66.99
(63.51, 70.47)
Incriminating
(violation)
United Nations 57.1
(53.48, 60.72)
70.6
(64.74, 76.46)
57.03
(50.13, 63.94)
44.86
(39.23, 50.49)
Military Court 59.4
(55.58, 63.21)
68.14
(61.66, 74.64)
58.78
(52.17, 65.4)
50.31
(42.91, 57.7)
Federal Court 59.2
(55.82, 62.57)
73.75
(67.62, 79.87)
63.37
(60.37, 70.36)
47.5
(41.88, 53.11)
Total (IJ) 58.56
(56.49, 60.62)
70.71
(67.2, 74.22)
60.64
(57.11, 64.16)
47.43
(43.93, 50.93)
Incriminating
(war crimes)
United Nations 52.05
(48.5, 55.59)
65.53
(59.31, 71.76)
54.54
(48.66, 60.42)
38
(32.04, 43.95)
Military Court 53.66
(49.9, 57.4)
64.82
(58.42, 71.23)
53.51
(47, 60.02)
44.41
(37.68, 51.14)
Federal Court 57.71
(54.35, 61.08)
70.79
(65.22, 76.35)
59.05
(53.99, 64.12)
41.96
(35.44, 48.49)
Total (WC) 54.47
(52.42, 56.52)
67.09
(63.63, 70.55)
55.91
(52.59, 59.22)
41.37
(37.7, 45.04)
Table reports the means of the Report Fairness, by Institution, Judgment and Ideology, with 95% confidence intervals in parentheses, N= 1,995.
Krebs The Legalization of Truth
84
(b) Experiment II:
Table 4. Summary of Treatment Effects: Mean Willingness to Prosecute the Soldiers, by Treatment Groups and Ideology
Treatment All
Respondents
Ideology
Liberals Moderates Conservatives
Legal Frame
Incriminating 30.25
(25.94, 34.55)
42.4
(32.81, 51.98)
26.66
(17.97, 35.35)
19.73
(14.12, 25.34)
Absolving 31.26
(26.81, 35.7)
36.27
(28.08, 44.45)
36.87
(29.1, 44.64)
18.73
(10.69, 26.76)
Moral Frame
Incriminating 39.43
(35.2, 43.66)
51.63
(42.76, 60.5)
40.88
(33.53, 48.24)
26.66
(20.5, 32.82)
Absolving 32.65
(28.01, 37.28)
50
(41.27, 58.72)
36.27
(28.32, 44.22)
15.27
(9.09, 21.46)
Control 30.54
(26.11, 34.97)
44.09
(34.76, 53.41)
32.75
(25.02, 40.49)
17.7
(10.63, 24.77)
Table reports the means of Willingness to Prosecute, by Treatment and Ideology, with 95% confidence intervals in parentheses, N= 998.